Abstract
Offline reinforcement learning (RL) has emerged as a promising paradigm for real-world applications since it aims to train policies directly from datasets of past interactions with the environment. The past few years, algorithms have been introduced to learn from high-dimensional observational states in offline settings. The general idea of these methods is to encode the environment into a latent space and train policies on top of this smaller representation. In this paper, we extend this general method to stochastic environments (i.e., where the reward function is stochastic) and consider a risk measure instead of the classical expected return. First, we show that, under some assumptions, it is equivalent to minimizing a risk measure in the latent space and in the natural space. Based on this result, we present Latent Offline Distributional Actor-Critic (LODAC), an algorithm which is able to train policies in high-dimensional stochastic and offline settings to minimize a given risk measure. Empirically, we show that using LODAC to minimize Conditional Value-at-Risk (CVaR) outperforms previous methods in terms of CVaR and return on stochastic environments.
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.Avoid common mistakes on your manuscript.
1 Introduction
Often, human decisions are stored and this builds interesting datasets. With the success of modern machine learning tools, comes the hope of exploiting them to construct useful decision helpers. To achieve this, we can use an imitation learning approach [1]. However, in this case, we will at best be as good as humans. Moreover, the performance of this approach depends heavily on the quality of the training dataset. In this work, we would like to avoid these limitations and thus, we consider another framework: reinforcement learning (RL).
RL has achieved impressive results in a number of challenging areas, including games [2, 3], robotic control [4, 5] and even health care [6, 7]. In particular, offline RL seems to be really interesting for real-world applications as it aims to train agents from a dataset of past interactions with the environment.
With the digitization of society, more and more features could be used to represent the environment. Unfortunately, classical RL algorithms are not able to work with high-dimensional states. Although we could manually choose a feature subset, this choice is not straightforward and it could have a huge impact on the performances. Therefore, it may be more practical to use RL algorithms capable of learning from high-dimensional states.
It is common to evaluate RL algorithms on deterministic (in the sense that the reward function is deterministic) environments such as the DeepMind Control suite [8]. However, in many real-world applications, environments are not deterministic but stochastic. In stochastic environments, the same action may lead to different outcomes due to randomness. This can make it difficult for the agent to distinguish between the quality of different actions and states, thereby complicating the learning process. Due to this inherent randomness, stochastic environments demand a larger number of samples to accurately estimate value functions and policies. On the other hand, offline learning has limited data available for training. The combination of these factors makes offline RL in stochastic environments particularly challenging. Furthermore, the performance of classical RL methods in such contexts remains unclear. Moreover, and as presented in Sect. 5.2, our experiments suggest that adding stochasticity to the environment decreases significantly the performance of these algorithms.
Based on these observations, we aim to develop a method which is able to train policies in offline settings with high-dimensional states while being robust to stochasticity. In this paper, we present Latent Offline Distributional Actor-Critic (LODAC), an algorithm designed to train policies in high-dimensional, stochastic environments within offline settings. The main idea is to encode the natural environment space into a smaller representation and train the agent directly in this latent space. But instead of considering the expected return, we use a risk measure. First, assuming some hypotheses, we show that minimizing this risk measure in the latent space is equivalent to minimizing the risk measure directly in the natural state. This theoretical result provides a natural framework: employ a latent variable model to encode the natural state into a latent space, and subsequently train a policy atop this latent space using a risk-sensitive RL algorithm. In the experimental part, we evaluate our algorithm on high-dimensional stochastic and deterministic datasets. In the best of our knowledge, we are the first authors to propose an algorithm to train policies in high-dimensional, stochastic and offline settings.
2 Related work
Before going further into our work, we present some related research.
As discussed and motivated in the previous section, this paper focuses on offline RL [9]. Offline RL is a specific approach of RL that aims to learn from past interactions with the environment. This framework holds promise for real-world applications as it enables the deployment of trained policies. Therefore, it is not surprising that offline RL has received a lot of attention in recent years [10,11,12,13,14,15]. One of the main problems in offline RL is that the Q-function tends to be overly optimistic in the case of out-of-distribution (OOD) states-actions pairs, where observational data is limited [16]. Several approaches have been introduced to address this issue. While some studies extend importance sampling methods [17, 18] other propose model-based techniques [19, 20] or even incorporate dataset re-weight sampling [21]. In our work, and following previous researches [22,23,24], we build a conservative estimation of the Q-function for OOD state-action pairs.
Another challenge encountered in our work is training policies from high-dimensional states. Over the past years, several algorithms have been proposed to address policy training within such contexts. For instance, several methods have been developed to directly handle image inputs [25,26,27,28]. Acquiring a meaningful representation of the observation is crucial to tackle this type of problem. Therefore, this issue has been studied by previous works [29, 30]. Some researchers employ data augmentation techniques to facilitate the learning of optimal representations [31]. However, these methods are limited to scenarios involving image states. Hence, in our work, we adhere to other common practices [32, 33] and we encode high-dimensional states into a latent space without relying on data augmentation. Furthermore, we integrate planning within this latent space, building upon studies that highlight its potential for performance enhancement [34,35,36]. Subsequently, a common approach involves training policies using classical RL algorithms, such as Soft Actor-Critic (SAC) [4], on top of this compact representation [37, 38]. Unfortunately, these classical RL algorithms struggle to train agents in stochastic environment. Therefore, in our study, we employ a risk-sensitive RL method atop this latent space.
Risk-sensitive RL is a specific approach to safe RL [39]. In safe RL, policies are trained to maximize performance while satisfying some safety constraints during training and/or in the deployment. In risk-sensitive RL, the focus is on minimizing a measure of the risk induced by the cumulative rewards, rather the maximizing the expected return. Risk-sensitive RL has gained attention in recent years [40,41,42,43]. Various risk measures may be considered, such that Exponential Utility [44] Cumulative Prospect Theory [45] or Conditional Value-at-Risk (CVaR) [46]. In this work, given the robust theoretical underpinnings and intuitive nature of Conditional Value-at-Risk [47, 48], we opt to focus on this operator. Furthermore, prior research suggests that taking into account Conditional Value-at-Risk instead of the classical expectation, can prevent the performance gap between simulations and real-world applications [49]. CVaR is really popular and has been strongly studied in the context of RL for many years [50,51,52]. For instance, [53] extend SAC in a distributional settings, enabling the training of risk-averse policies. However, this algorithm is not suited for offline RL. Other authors presented O-RAAC [54], an algorithm which is able to train risk-averse policies in an offline settings. In this paper, OOD errors are managed through an imitation learning component [55]. Finally, Ma et al. [56] presented a method for training risk-averse policies in an offline setting. In our work, we aim to extend this method to enable the training of policies in the context of high-dimensional states.
3 Preliminaries
In this section, we introduce notations and recall concepts we will use later.
Coherent risk measure Let \((\Omega , {\mathcal {F}}, {\mathbb {P}})\) a probability space and \({\mathcal {L}}^{2}:= {\mathcal {L}}^{2}(\Omega , {\mathcal {F}}, {\mathbb {P}})\). A function \({\mathcal {R}} \text {: } {\mathcal {L}}^{2} \rightarrow (-\infty , +\infty ]\) is called a coherent risk measure [57] if
-
1.
\({\mathcal {R}}(C) = C \text { for all constants } C\).
-
2.
\({\mathcal {R}}((1 - \lambda )X + \lambda X) \le (1 - \lambda ) {\mathcal {R}}(X) + \lambda {\mathcal {R}}(X) \text { for } \lambda \in (0, 1)\).
-
3.
\({\mathcal {R}}(X) \le {\mathcal {R}}(X')\text { when } X \le X'\).
-
4.
\({\mathcal {R}}(X) \le 0 \text { when } \Vert X_{k} - X \Vert _{2} \rightarrow 0 \text { with } {\mathcal {R}}(X_{k}) \le 0\).
-
5.
\({\mathcal {R}}(\lambda X) = \lambda {\mathcal {R}}(X) \text { for } \lambda > 0\).
Coherent risk measures have some interesting properties. In particular, a risk measure is coherent if and only if, there exists a risk envelope \({\mathcal {U}}\) such that
[48, 58, 59]. A risk envelope is a nonempty convex subset of \({\mathcal {P}}\) that is closed and where \({\mathcal {P}}:= \{ \delta \in {\mathcal {L}}^{2} \text { } | \text { } \text { } \delta \ge 0 \text {, } {\mathbb {E}}_{p}[\delta ] = 1 \}\).
The definition of a risk measure \({\mathcal {R}}\) might depend on a probability distribution p. In some cases, it may be useful to specify which distribution we are working with. Thus, we sometimes use the notation \({\mathcal {R}}_{p}\). There exists a lot of different coherent risk measures, for example, the Wang risk measure [60], the entropic Value-at-Risk [61] or Conditional Value-at-Risk [46, 62].
Conditional Value-at-Risk (CVaR\(_{\alpha }\)) with probability level \(\alpha \in (0, 1)\) is defined as
Moreover the risk envelope associated to CVaR\(_{\alpha }\), can be written as \({\mathcal {U}} = \{ \delta \in {\mathcal {P}} \text { } | \text { } {\mathbb {E}}_{p}[\delta ] = 1 \text {, } 0 \le \delta \le \frac{1}{\alpha } \}\) [59, 63]. This rigorous definition may not be intuitive, but roughly speaking, CVaR\(_{\alpha }\) is the expected value of X given the upper \(\alpha\)-tail of its conditional distribution, representing the \((1-\alpha )\) worst-case scenarios. In this paper, we use the classical definition of Conditional Value-at-Risk presented in risk measure literature. In particular, X should be interpreted as a loss function.
Offline RL We consider a Markov Decision Process (MDP), \((S, {\mathcal {A}}, p, r, \mu _{0}, \gamma )\) where S is the environment space, \({\mathcal {A}}\) the action space, r the reward distribution (\(r_{t} \sim r(\cdot | s_{t}, a_{t})\)), p the transition probability distribution (\(s_{t+1} \sim p( \cdot | s_{t}, a_{t}))\). \(\mu _{0}\) is the initial state distribution and \(\gamma \in (0, 1)\) denotes the discount factor. For the purpose of notation, we define \(p(s_{0}):= \mu _{0}(s_{0})\) and we write the MDP \((S, {\mathcal {A}}, p, r, \gamma )\).
Actions are taking following a policy \(\pi\) which depends on the environment state, (i.e., \(a_{t} \sim \pi (\cdot | s_{t})\)). A sequence on the MDP \((S, {\mathcal {A}}, p, r, \gamma )\), \(\tau = s_{0}, a_{0}, r_{0}, \ldots s_{H}\), with \(s_{i} \in S\) and \(a_{i} \in {\mathcal {A}}\) is called a trajectory. A trajectory of fixed length \(H \in {\mathbb {N}}\) is called an episode. Given a policy \(\pi\), a rollout from state-action \((s, a) \in S \times {\mathcal {A}}\) is a random sequence \(\{ (s_{0}, a_{0}, r_{0}), (s_{1}, a_{1}, r_{1}), \ldots \}\) where \(a_{0}=a\), \(s_{0}=s\), \(s_{t+1} \sim p( \cdot | s_{t}, a_{t})\), \(r_{t} \sim r(\cdot | s_{t}, a_{t})\) and \(a_{t} \sim \pi (\cdot | s_{t})\). Given a policy \(\pi\) and a fixed length H, we have a trajectory distribution given by
The goal of classical risk-neutral RL algorithms is to find a policy that maximizes the expected discounted return \({\mathbb {E}}_{\pi }[\sum _{t=0}^{H} \gamma ^{t}r_{t}]\) where H could be infinite. Equivalently, we can look for the policy that maximizes the Q-function, which is defined as \(Q_{\pi } \text {: } S \times {\mathcal {A}} \rightarrow {\mathbb {R}}\), \(Q_{\pi }(s_{t}, a_{t}):= {\mathbb {E}}_{\pi }[\sum _{t'=t}^{H} \gamma ^{t'-t} r_{t'}]\).
Instead of this classical objective function, other choices are possible. For example, the maximum entropy RL objective function \({\mathbb {E}}_{\pi }[ \sum _{t=0}^{H} r_{t} + {\mathcal {H}}(\pi ( \cdot | s_{t}))]\), where \({\mathcal {H}}\) denotes the entropy. This function has an interesting connection with variational inference [64] and it has shown impressive results in recent years [4, 19, 65].
In offline RL, we have access to a fixed dataset \({\mathcal {D}} = \{ (s_{t}, a_{t}, r_{t}, s_{t+1}) \}\), where \(s_{t} \in S\), \(a_{t} \in {\mathcal {A}}\), \(r_{t} \sim r(s_{t}, a_{t})\) and \(s_{t+1} \sim p(\cdot | s_{t}, a_{t})\), and we aim to train policies without interaction with the environment. A such dataset comes with the empirical behavior policy \(\pi _{\beta }(a | s):= \frac{\sum _{(s_{t}, a_{t}) \in {\mathcal {D}}} \mathbbm {1}_{\{ s_{t}=s, a_{t}=a \}}}{\sum _{s_{t} \in {\mathcal {D}}} \mathbbm {1}_{\{s_{t}=s \}}}.\)
Latent variable model There are different methods for learning directly from high-dimensional states [29, 65, 66]. However, in this work, we build upon the framework presented in Stochastic Latent Actor-Critic (SLAC) [38]. The main idea of this work is to train a latent variable model to encode the natural MDP \((S, {\mathcal {A}}, p, r, \gamma )\) into a latent space \((Z, {\mathcal {A}}, q, r, \gamma )\) and to train policies directly in this space. To achieve this, the variational distribution \(q(z_{1:H}, a_{t+1:H} | s_{1:t}, a_{1:t})\) is factorized into a product of inference term \(q(z_{i+1} | z_{i}, s_{i+1}, a_{i})\), latent dynamic term \(q(z_{i+1} | z_{i}, a_{i})\) and policy term \(\pi (a_{i} | s_{1:t}, a_{1:t-1})\) as follows
Using this factorization, the evidence lower bound (ELBO) [67] and a really interesting theoretical approach [64], the following objective function for the latent variable model is derived
where \(D_{KL}\) is the Kullback–Leibler divergence and D is a decoder.
Distributional RL The aim of distributional RL is to learn the distribution of the discounted cumulative rewards \(Z_{\pi }:= - \sum _{t}^{H} \gamma ^{t}r_{t}\). \(Z_{\pi }\) is a random variable. A classical approach to distributional RL is to learn \(Z_{\pi }\) implicitly using its quantile function \(F^{-1}_{Z(s, a)} \text {: } [0, 1] \rightarrow {\mathbb {R}}\), which is defined as \(F^{-1}_{Z(s, a)}(y):= \inf \{ x \in {\mathbb {R}} \text { } | \text { } y \le F_{Z(s, a)}(x) \}\) and where \(F_{Z(s, a)}\) is the cumulative density function of the random variable Z(s, a). A model \(Q_{\theta }(\eta , s, a)\) is used to approximate \(F^{-1}_{Z(s, a)}(\eta )\). For \((s, a, r, s') \sim {\mathcal {D}}\), \(a' \sim \pi (\cdot | s')\), \(\eta , \eta ' \sim \text {Uniform}[0, 1]\), we define \(\delta := r + \gamma Q_{\theta '}(\eta ', s', a') - Q_{\theta }(\eta , s, a)\). \(Q_{\theta }\) is trained using the \(\tau\)-Huber quantile regression loss at threshold k [68]
With this function \(F_{Z(s, a)}\), different risk measures can be computed, such that Cumulative Probability Weight (CPW) [45], Wang measure [60] or Conditional Value-at-Risk. For example, the following equation [69] is used to compute CVaR\(_{\alpha }\)
It is also possible to extend distributional RL to offline settings. For example, O-RAAC [54] decomposes the actor into two different components an imitation actor and a perturbation model. CODAC [56] extends DSAC [70] in offline settings. More precisely, the \(Q_{\theta }(\eta , s, a)\) are trained using the following loss function
where \(U = \text {Uniform}[0, 1]\). The first term of the equation, is introduced to avoid overly optimistic estimations for OOD state-action pairs [22]. The second term (i.e. \({\mathcal {L}}_{k}(\delta , \tau ))\) is the classical objective function used to train the Q-function in a distributional settings.
4 Theoretical considerations
In this paper, we aim to train policies in high-dimensional, stochastic environment within offline settings. For a practical point of view, our idea is straightforward. Since training with high-dimensional states directly fails and following previous works [19, 38], we encode our high-dimensional states into a more compact representation using \(\phi \text {: } S \rightarrow Z\), where \(Z = \phi (S)\). Then, using \(\phi\), we build an MDP in the latent space and train policies directly on top of this space.
However, from a theoretical point of view, this general idea is not entirely clear. Is this latent MDP always well-defined? If we find a policy that minimizes a risk measure in the latent space, will it also minimize the risk measure in the natural state?
The first goal of this section is to rigorously construct an MDP in the latent space. Then, we show that, for certain coherent risk measures and under some assumptions, minimizing the risk measure in the latent space is equivalent to minimizing it in the natural space. In particular, we demonstrate that it is the case for Conditional Value-at-Risk.
4.1 Theoretical results
First, we make the following assumptions
-
1.
\(\forall a \in {\mathcal {A}}\) we have \(r(\cdot | s, a) = r(\cdot | s', a)\) if \(\phi (s) = \phi (s')\).
-
2.
We note \({\mathbb {P}}(\cdot | s_{t}, a_{t})\) the probability measure with probability density function \(p(\cdot | s_{t}, a_{t})\). We suppose that if \(s_{t}, s'_{t} \in S\) satisfy \(\phi (s_{t}) = \phi (s'_{t})\), then \({\mathbb {P}}(\cdot | s_{t}, a_{t}) = {\mathbb {P}}(\cdot | s_{t}', a_{t})\).
-
3.
We note \({\mathbb {Q}}(\cdot | z_{t}, a_{t})\) the probability image of \({\mathbb {P}}(\cdot | s_{t}, a_{t})\) by \(\phi\) and where \(s_{t}\) is any element of \(\phi ^{-1}(z_{t})\). We suppose \({\mathbb {Q}}(\cdot | z_{t}, a_{t})\) admits a probability density functions \(q(\cdot | z_{t}, a_{t})\).
Under these assumptions, \(\phi\) induces an MDP \((Z, {\mathcal {A}}, q, r', \gamma )\) where the reward distribution \(r'\) is defined as \(r'(\cdot | z, a):= r(\cdot | s, a)\) for any \(s \in \phi ^{-1}(z)\). \(r'\) is well defined by surjectivity of \(\phi\) and by assumption (1). \({\mathbb {Q}}\) is well defined by hypothesis (2).
Obviously, for a given policy \(\pi '\) on the MDP \((Z, {\mathcal {A}}, q, r', \gamma )\) and fixed length H, we have a trajectory distribution
Given a policy \(\pi\) and a fixed length H, we denote \(\Omega\) the set of all trajectories on the MDP \((S, {\mathcal {A}}, p, r, \gamma )\), \({\mathcal {F}}\) the \(\sigma\)-algebra generated by these trajectories and \({\mathbb {P}}_{\pi }\) the probability with probability distribution \(p_{\pi }\). Following the same idea and given a policy \(\pi '\), we denote \(\Omega '\) the set of trajectories of length H on the MDP \((Z, {\mathcal {A}}, q, r')\), \(\mathcal {F'}\) the \(\sigma\)-algebra generated by these trajectories and \({\mathbb {Q}}_{\pi '}\) the probability with probability distribution \(q_{\pi '}\). \((\Omega , {\mathcal {F}}, {\mathbb {P}}_{\pi })\) and \((\Omega ', \mathcal {F'}, {\mathbb {Q}}_{\pi '})\) are probability spaces.
\(\phi \text {: } S \rightarrow Z\) induces a map
With a slight abuse of notation, we write it \(\phi\).
We denote \(\Pi\) the set of all policies \(\pi\) on the MDP \((S, {\mathcal {A}}, p, r)\) which satisfies \(\pi (a | s) = \pi (a | s')\) if \(\phi (s) = \phi (s')\). If \(\pi \in \Pi\), then \(\pi\) induces a policy \(\pi '\) on the MDP \((Z, {\mathcal {A}}, q, r')\), taking \(\pi '(a | z):= \pi (a | s)\) where s is any element of \(\phi ^{-1}(z)\).
For any \(\pi \in \Pi\), we denote \(\pi '\) the associated policy on the MDP \((Z, {\mathcal {A}}, q, r')\) as defined above. Furthermore, we note \(\Pi ':= \{ \pi ' \text { } | \text { } \pi \in \Pi \}\). If \(\pi ' \in \Pi '\), a policy \(\pi \in \Pi\) can be defined as \(\pi (a | s):= \pi '(a | \phi (s))\). X is a random variable on \(\Omega\) and \(X'\) a random variable on \(\Omega '\).
With all these notations, we can introduce our first result.
Lemma 1
Let \(\pi \in \Pi\). Then, \({\mathbb {Q}}_{\pi '}\) is the probability image of \({\mathbb {P}}_{\pi }\) by \(\phi\).
Proof
Let \(B':= Z_{0} \times A_{0} \times R_{0} \times \ldots Z_{H} \in {\mathcal {F}}'\). We have
Then, denoting \({\bar{s}}_{0}\) one arbitrary element of \(\phi ^{-1}(z_{0})\), we obtain
Iterating this process, we find
\(\square\)
This result is not a surprise but it has an interesting implication. Indeed, it implies that if \(X' \circ \phi = X\), then \({\mathbb {E}}_{q_{\pi '}}[X'] = {\mathbb {E}}_{p_{\pi }}[X]\). And thus, we get the following result.
Corollary 1
Suppose \(X' \circ \phi = X\). Then, if a policy \(\pi _{\star }'\) satisfies
its associated policy \(\pi _{\star }\) satisfies
For a coherent risk measure \({\mathcal {R}}\), we write \({\mathcal {U}}\) the risk envelope associated to \({\mathcal {R}}\) in \(\Omega\) and \(\mathcal {U'}\) the corresponding risk envelope in \(\Omega '\). If \(\forall \delta \in {\mathcal {U}}\), \(\exists \delta ' \in \mathcal {U'}\) such that \(\delta = \delta ' \circ \phi\) and \(\forall \delta ' \in \mathcal {U'}\) \(\exists \delta \in {\mathcal {U}}\) with \(\delta = \delta ' \circ \phi\) almost everywhere, we write \({\mathcal {U}} = \mathcal {U'} \circ \phi\).
Proposition 1
Let \({\mathcal {R}}\) be a coherent risk measure. Suppose that \(X' \circ {\phi } = X\) and \({\mathcal {U}} = \mathcal {U'} \circ \phi\). Then, if a policy \(\pi _{\star }'\) satisfies
its associated policy \(\pi _{\star }\) verifies
Proof
Since \({\mathcal {U}} = \mathcal {U'} \circ \phi\), we have
Thus
Now, let \(\pi _{\star }':= \text {argmin}_{\pi ' \in \Pi } {\mathcal {R}}(X')\) and \(\pi _{\star }\) the associated policy. By contradiction, suppose there exists \(\pi _{1}\) with \({\mathcal {R}}_{\pi _{1}}(X) < {\mathcal {R}}_{\pi _{\star }}(X)\). But in this case and by the above observation, we would have \({\mathcal {R}}_{\pi _{1}'}(X') < {\mathcal {R}}_{\pi _{\star }'}(X')\). \(\square\)
This result provides a useful criteria to prove the optimality of a method. For example, consider \({\mathcal {R}}(X) = {\mathbb {E}}_{p_{\pi }}[X]\) which is obviously a coherent risk-measure. Its risk envelope is \({\mathcal {U}} \equiv 1\). And thus, Corollary 1 follows directly from the last proposition too.
We also have the same result in the case where \({\mathcal {R}}(X) = \text {CVaR}_{\alpha }(X)\).
Proposition 2
Suppose that \(X' \circ {\phi } = X\). Then, if a policy \(\pi _{\star }'\) satisfies
its associated policy \(\pi _{\star }\) verifies
Proof
Recall that the risk envelop of the CVaR\(_{\alpha }\), takes the form \({\mathcal {U}} = \{ \delta \text { } | \text { } 0 \le \delta \le \frac{1}{\alpha } \text { } {\mathbb {E}}_{p}[\delta ] = 1 \}\). First, we will show that for each \(\delta ' \in {\mathcal {U}}'\) there exists \(\delta \in {\mathcal {U}}\) such that \({\mathbb {E}}_{q}[\delta 'X'] = {\mathbb {E}}_{p}[\delta X]\). Then, we will show that for each \(\delta \in {\mathcal {U}}\) there exists \(\delta ' \in {\mathcal {U}}'\) such that \({\mathbb {E}}_{p}[\delta X] = {\mathbb {E}}_{q}[\delta 'X']\).
-
Let \(\delta ' \in \mathcal {U'}\). We define \(\delta := \delta ' \circ \phi\). By construction, we have \(0 \le \delta \le \frac{1}{\alpha }\) and \(\begin{aligned} {\mathbb {E}}_{p}[\delta ] = {\mathbb {E}}_{p}[\delta ' \circ \phi ] = {\mathbb {E}}_{q}[\delta '] = 1 \end{aligned}\). Thus, \(\delta \in {\mathcal {U}}\). With the same idea, remark that \({\mathbb {E}}_{q}[\delta 'X'] = {\mathbb {E}}_{p}[\delta X]\).
-
Let \(\delta \in {\mathcal {U}}\). By construction, \(\delta p_{\pi }\) is a density function. Let \(\tilde{{\mathbb {P}}}_{\pi }\) its associated probability measure. We consider \(\tilde{{\mathbb {Q}}}_{\pi '}\) the probability image of \(\tilde{{\mathbb {P}}}_{\pi }\) by \(\phi\). Then, remark that \(\tilde{{\mathbb {Q}}}_{\pi '}\) is absolutely continuous with respect to \({\mathbb {Q}}_{\pi '}\). Indeed, if A satisfies \({\mathbb {Q}}_{\pi '}(A) = 0\), we have
$$\begin{aligned} \tilde{{\mathbb {Q}}}_{\pi '}(A)&= \tilde{{\mathbb {P}}}_{\pi }(\phi ^{-1}(A)) = \int _{\phi ^{-1}(A)} \delta (\tau )p(\tau )\textrm{d}\tau \\&\le \frac{1}{\alpha } {\mathbb {P}}_{\pi }(\phi ^{-1}(A)) = \frac{1}{\alpha } {\mathbb {Q}}_{\pi }(A) = 0 \end{aligned}$$Thus by Radon-Nikodym, there exists \(\delta ' \ge 0\), such that for all \(B \in {\mathcal {F}}'\)
$$\begin{aligned} \tilde{{\mathbb {Q}}}_{\pi '}(B) = \int _{B} \delta ' \textrm{d}{\mathbb {Q}}_{\pi '} = \int _{B} \delta '(\tau ') q_{\pi '}(\tau ') \textrm{d}\tau ' \end{aligned}$$We will show that \(\delta ' \in {\mathcal {U}}'\). We define \(C:= \{ \tau ' \in \Omega ' \text { } | \text { } \delta '(\tau ) > \frac{1}{\alpha } \}\). By contradiction, suppose that C is not empty. On one hand have
$$\begin{aligned} \tilde{{\mathbb {Q}}}_{\pi '}(C)&= \int _{C} \delta '(\tau ')q_{\pi }(\tau ')\textrm{d}\tau ' \\&> \frac{1}{\alpha } \int _{C} q_{\pi }(\tau ')\textrm{d}\tau ' = \frac{1}{\alpha } {\mathbb {Q}}_{\pi '}(C) \end{aligned}$$And on the other hand, we have
$$\begin{aligned} \tilde{{\mathbb {Q}}}_{\pi }(C)&= \tilde{{\mathbb {P}}}_{\pi }(\phi ^{-1}(C)) \\&= \int _{\phi ^{-1}(C)} \delta (\tau ) p_{\pi }(\tau ) \textrm{d}\tau \\&\le \frac{1}{\alpha } {\mathbb {P}}_{\pi }(\phi ^{-1}(C)) = \frac{1}{\alpha } {\mathbb {Q}}_{\pi '}(C) \end{aligned}$$Therefore C is empty and thus, \(0 \le \delta ' \le \frac{1}{\alpha }\). Then, since \(\delta 'q_{\pi '}\) is a probability density function \(\delta ' \in {\mathcal {U}}'\).
Finally, since, \(\delta 'q_{\pi '}\) is the probability density function of the probability image of \(\tilde{{\mathbb {P}}}_{\pi }\), we obtain \({\mathbb {E}}_{\delta p_{\pi }}[X] = {\mathbb {E}}_{\delta ' q_{\pi '}}[X']\)
\(\square\)
In particular, for
and
(with \(\beta\) potentially null) we have
Therefore, we get the following result.
Corollary 2
Let \(X \text {: } \Omega \rightarrow {\mathbb {R}}\) and \(X' \text {: } \Omega ' \rightarrow {\mathbb {R}}\) defined as \(X(\tau ): = \sum _{t=0}^{H} -r_{t} - \beta {\mathcal {H}}(\pi (\cdot | s_{t}))\) and \(X'(\tau '):= \sum _{t=0}^{H} - r_{t} - \beta {\mathcal {H}}(\pi '(\cdot | z_{t}))\). Then, if a policy \(\pi _{\star }'\) satisfies
its associated policy \(\pi _{\star }\) verifies
4.2 Discussion
The results presented in the last section establish a theoretical equivalence between minimizing the risk measure in the latent space and in the natural space. However, to obtain this guarantee we need to make some assumptions.
Assumptions (1), (2) ensure that the reward distribution \(r(\cdot | s_{t}, a_{t})\) and the probability measure \({\mathbb {P}}(\cdot | s_{t}, a_{t})\) are insensitive to any change in \(\phi ^{-1}(z_{t})\). Moreover \(\pi \in \Pi\) ensures that the policy distribution is stable to any change in \(\phi ^{-1}(z_{t})\). Thus, and roughly speaking, these assumptions guarantee we do not lose information by encoding \(s_{t}\) into \(\phi (s_{t})\), in terms of reward distribution, transition probability measure and the optimal policy.
These theoretical considerations highlight that in order to learn a meaningful latent representation of the natural MDP, we should consider all components of the MDP and not only focus on the environment space.
5 Latent Offline Distributional Actor-Critic
The theoretical results presented above justify our really natural idea: encode the natural environment space S into a compact representation Z and then, use a risk-sensitive offline RL algorithm to learn on top of this space. This is the general idea of LODAC.
5.1 Practical implementations of LODAC
In this section, we present the practical implementation of LODAC used in our experiments.
First, we need to implement and train a latent variational latent model. Following previous works [19, 23], our latent variable model contains the following components
The image encoder, denoted as \(E_{\theta }\), is a classical Convolutional Neural Network consisting of 4 2D convolutional layers. Each of them uses a kernel of size 4, a stride of 2 and Rectified Linear Unit (ReLU) as activation function. These layers use 32, 64, 128 and 256 filters, respectively.
The inference and the latent transition model are implemented as a Recurrent State Space Model (RSSM) [66]. Specifically, a latent state contains two components \(z_{t} = [d_{t}, x_{t}]\) where \(d_{t}\) is the deterministic part and \(x_{t}\) the stochastic. \(d_{t}\) is computed using a dense layer which takes \(x_{t-1}, a_{t-1}\) as input, uses Exponential Linear Unit as activation function and contains 256 hidden units. This layer is followed by a gated recurrent unit (GRU) cell which uses the hyperbolic tangent as activation function. The dimension of the output is 256. \(d_{t}\) corresponds to the hidden state of this GRU cell.
Then, the latent transition model employs two dense layers. The first one uses Exponential Linear Unit as activation function and yields an output of dimension 256. The second one does not have any activation function and provide an output of dimension 128. This output is subsequently split into two parts. The first part is interpreted as the mean of a normal distribution, while the second part (after applying a softplus) represents the standard deviation. The stochastic component of the latent state \(x_{t}\) is sampled from this distribution.
The inference model computes the deterministic part of the latent state \(d_{t}\) using the first layers of the latent transition model. Then, \(d_{t}\) and \(h_{t}\) are fed into two consecutive dense layers. The first one uses an Exponential Linear Unit activation function and 256 hidden units. The final layer does not have any activation function and the dimension of the output is 128. Similar to the latent transition model, the output is divided into two parts, which are interpreted as the mean and the variance of a normal distribution. \(x_{t}\) is sampled from this distribution.
Finally, the decoder \(D_{\theta }\) consists of a dense layer with 1024 units followed by four deconvolutional layers. Each of them have a stride of 2 and except the last layer which do not have any activation function. All layers have ReLU as activation function. The first two deconvolutional layers have a kernel size of 5, while the last two have a kernel size of 6. Finally, these layers contain 128, 64, 32, 32 and 3 filters, respectively. The reconstructed image is sampled from a normal distribution where the mean is determined by \(D_{\theta }\) and the standard deviation is fixed to one. The main components of our latent variable model are presented in Fig. 1.
These models have been trained using Adam optimizer, a batch size of 64, a learning rate of \(6*10^{-4}\) and the following objective function [19]
Rewards estimation come from a sampling from a normal distribution where the mean is determined by the output of a neural network \(r_{\theta }\) and the standard deviation is fixed to one. \(r_{\theta }\) contains one hidden dense layer with 128 units and uses Exponential Linear Unit as activation function. \(r_{\theta }\) is trained using maximum log-likelihood.
After the training of these models, the dataset \({\mathcal {D}}\) is encoded into the latent space and stored in a replay buffer \({\mathcal {B}}_{\text {latent}}\). More precisely, \({\mathcal {B}}_{\text {latent}}\) contains transitions of the form \((z_{1:H}, r_{1:H}, a_{1:H})\) where \(z_{1:H} \sim \phi _{\theta }(\cdot | s_{1:H}, a_{H-1})\) and \(s_{1:H}, r_{1:H-1}, a_{1:H-1} \sim {\mathcal {D}}\). Next, we introduce a latent buffer \({\mathcal {B}}_{\text {synthetic}}\) which contains rollouts transitions performed using the policy \(\pi _{\theta }\), the latent model \(q_{\theta }\) and the reward estimator \(r_{\theta }\).
The policy \(\pi _{\theta }\) comprises three dense layers. The first two use ReLU as activation function and 256 hidden units. The final layer does not use any activation function and yields two distinct numbers. The first number is interpreted as the mean of a normal distribution and the second one as a standard deviation. Action is sampled from this distribution.
The critic \(Q_{\theta }\) is implemented as a quantile distributional critic network [71]. Specifically, we have \(Q_{\theta }(\eta , z, a) = f_{\theta }(\psi _{\theta }(z, a) \odot \xi _{\theta }(\eta ))\) where \(\odot\) denotes the element-wise (Hadamard) product. \(\psi _{\theta }\) is a dense layer with ReLU as activation function, provides an output of dimension 256 and uses layer normalization [72]. \(\xi _{\theta }\) consists of an embedding of \(\eta\) into a space of dimension 64 followed by a dense layer. Specifically, \(\eta\) is embedded into a vector where the components take the form \(\cos (i \eta \pi )\) with i ranging from 1 to 64. After that, this embedded vector is fed to a dense layer which uses sigmoid as activation function and produces an output of dimension 256. This layer also employs layer normalization. Finally, the model \(f_{\theta }\) consists of two dense layers. The first one uses layer normalization, ReLU as activation function and 256 hidden units. The last layer does not use any activation function and produces an output of dimension one.
Actor \(\pi _{\theta }\) and critic \(Q_{\theta }\) are trained on \({\mathcal {B}}:= {\mathcal {B}}_{\text {synthetic}} \cup {\mathcal {B}}_{\text {latent}}\). To achieve this, and based on our empirical results, we follow Ma et al. [56]. Thus, the critic \(Q_{\theta }\) is iteratively chosen to minimize
where \(U = \text {Uniform}[0, 1]\),
with \((z_{t}, a_{t}, r_{t}) \sim {\mathcal {B}}\), \(a_{t+1} \sim \pi _{\theta }(\cdot | z_{t})\) and \({\mathcal {L}}_{k}\) is the \(\tau\)-Huber quantile regression loss as defined in (5). Following Kumar et al. [22], Ma et al. [56], we add two parameters \(\zeta , \omega \in {\mathbb {R}}_{> 0}\) to the last equation
\(\zeta\) is used to threshold the difference between \({\mathbb {E}}_{(z, a) \sim {\mathcal {B}}} \left[ Q_{\theta }(\eta , z, a) \right]\) and the regularizer \({\mathbb {E}}_{ z \sim {\mathcal {B}}} \left[ \log \sum _{a} \exp (Q_{\theta }(\eta , z, a)) \right]\). The parameter \(\omega\) scales this difference. Remark that if this difference (scaled to \(\omega\)) is smaller than the parameter \(\zeta\), then \(\alpha\) will be set to 0 and only the term \({\mathcal {L}}_{k}(\delta , \eta )\) will be considered.
Furthermore, since the action space \({\mathcal {A}}\) is continuous, the computation \(\log \sum _{a} \exp (Q_{\theta }(\eta , z, a)\) is intractable. To overcome this problem, and as introduced in Kumar et al. [22], we use the following approximation
where \(U({\mathcal {A}}) = \text {Unif}({\mathcal {A}})\) and we choose \(M=10\).
The actor is trained to minimize Conditional Value-at-Risk of the negative cumulative reward, which can be computed using the formula
\(\pi _{\theta }\) and \(Q_{\theta }\) have been trained using Adam optimizer with a batch size of 256. Following previous works [19, 23], batches of equal data mixed from \({\mathcal {B}}_{\text {latent}}\) and \({\mathcal {B}}_{\text {synthetic}}\) are used. While the critic has been trained using a learning rate of \(3*10^{-4}\), we used a value of \(3*10^{-5}\) for the actor. Finally, it is worth mentioning we used the clipped double Q-learning trick [73] for training \(Q_{\theta }\). A summary of LODAC can be found in Algorithm 1.
Remark that for training a policy using the approach presented above, we need to be able to compute an expectation of a uniform distribution (Eq. 9), approximate \(\log \sum _{a} \exp (Q_{\theta }(\eta , z, a))\) using 10 and estimate Conditional Value-at-Risk using the formula 11 to compute the loss of the policy. Unfortunately, all these computations take time and are unavoidable. Thus, it is not surprising that optimizing a policy using this approach takes more time than classical methods.
![figure a](http://media.springernature.com/lw685/springer-static/image/art%3A10.1007%2Fs00521-023-09029-3/MediaObjects/521_2023_9029_Figa_HTML.png)
5.2 Experimental setup
In this section, we evaluate the performance of LODAC.
First, our method is compared with LOMPO [19], an offline high-dimensional risk-free algorithm. Then, we build a version of LODAC where the actor and the critic are trained using O-RAAC [54]. We denote it LODAC-O. Moreover, it is also possible to use a risk-free offline RL in the latent space. Thus, a risk-free policy is also trained in the latent space using COMBO [23].
These algorithms are evaluated on the standard walker walk task from the DeepMind Control suite [8], but here we learn directly from the pixels of dimensions \(3\times 64 \times 64\). As a standard practice, each action is repeated two times on the ground environment and episodes of length 1000 are used. These algorithms are tested on three different datasets: expert, medium and expert-replay. Each dataset consists of 100K transitions steps. More precisely, the following datasets are built.
-
Expert For the expert dataset, actions are chosen according to an expert policy that has been trained online, in a risk-free environment using SAC for 500K training steps. The states used for this training are the classical states provided by DeepMind Control suite.
-
Medium In this dataset, actions are chosen according to a policy that has been trained using the same method as above. However, here, the training was stopped when the policy achieves about half the performance of the expert policy.
-
Expert_replay The expert_replay dataset consists of episode that were sampled from the expert policy during the training.
The setup presented above allows to test our approach on deterministic environments.
However, we would like to evaluate our algorithm on stochastic environments too. To achieve this, we use the same datasets, but we modify the reward using the following formula
where r is the classical reward function. \({\overline{r}}\), \(\lambda\) are hyperparameters and \({\mathcal {B}}_{p_{0}}\) is a Bernoulli distribution of parameter \(p_{0}\). In our experiments, we choose \(\lambda =8\) and \(p_{0}=0.1\). Different values of \({\overline{r}}\) are used for each dataset such that about the half of the states verify \(r(s_{t}, a_{t}) > {\overline{r}}\).
Hyperparameters of each model have been manually tuned based on our experiments and hyperparameters used in previous papers. However, due to the substantial time required to train these models (specifically LODAC and LODAC-O), we were unable to test a wide range of hyperparameter combinations.
Algorithms have been tested using the following procedure. First of all, to avoid excessive computation time and for a more accurate comparison, we use the same latent variable model for all algorithms.
We evaluate each algorithm using 100 episodes, reporting the mean and CVaR\(_{\alpha }\) of the returns. LODAC and LODAC-O are trained to minimize CVaR\(_{0.7}\). LOMPO and COMBO are trained to maximize the return. We run 4 different random seeds. As introduced in Fu et al. [74], we use the normalized score to compare our algorithms. Specifically, a score of 0 corresponds to a fully random policy and a score of 100 corresponds to an expert policy on the deterministic task. However, and as suggested in Agarwal et al. [75], instead of taking the mean of the results, we consider the interquartile means (IQM).
5.3 Results discussion
In this section, we discuss the results of our experiments, which are presented in Table 1 for the stochastic environment and in Table 2 for the deterministic environment. We bold the highest score across all methods. Complete results of all different runs can be found in Appendix A.
Stochastic environment The first general observation is that LODAC and LODAC-O generally outperform risk-free algorithms. The only exception is with the expert_replay dataset where LODAC-O provides worse results. This is not really surprising since actors trained with O-RAAC contains an imitation component, and obviously, the imitation agent provides really poor performance on this dataset. LODAC provides really interesting results as it significantly outperforms risk-free algorithm in terms of \(\text {CVaR}_{0.7}\) and return on all datasets. Moreover, it provides better results than LODAC-O on the medium and the expert_replay dataset, while achieving comparable result on the expert dataset. A final observation is that risk-free RL policies provide generally bad performances on this stochastic environment.
Deterministic environment First, a drop of performance between the deterministic and the stochastic environment can be noticed. This is not surprising, as stochastic environments are generally more challenging than deterministic ones. However, this change appears to affect risk-free algorithms more significantly than LODAC-O or LODAC. Indeed, we get a difference of more than \(27\%\) with LOMPO and COMBO on the medium and expert_replay dataset in terms of return between the deterministic and the stochastic environment. This difference even reached \(42\%\) for COMBO on the expert_replay dataset. For LODAC-O and for the same tasks, we obtain a deterioration of less than \(17\%\). In contrast to risk-free methods, and in a lesser degree LODAC-O, adding stochasticity to the dataset does not seem to have a significant impact on the performance of LODAC. Indeed, with this algorithm we observe a drop in performance of less than \(9 \%\). For the medium dataset, a difference of only \(5.23\%\) can even be noticed.
6 Conclusion
While offline RL appears to be an interesting paradigm for real-world applications, many of these real-world applications are high-dimensional and stochastic. However, current high-dimensional offline RL algorithms are trained and tested in deterministic environments. Our empirical results suggest that adding stochasticity to the training dataset significantly decreases the performance of high-dimensional risk-free offline RL algorithms.
Based on this observation, we develop LODAC. LODAC can be used to minimize various risk measures like Conditional Value-at-Risk. Our theoretical considerations in Sect. 4 show that our algorithm relies on a strong theoretical foundation. Finally, the use of LODAC to minimize CVaR empirically outperforms previous algorithms in term of CVaR and return on stochastic high-dimensional environments.
Data availability statement
The datasets generated during the current study are available from the corresponding author on reasonable request.
References
Hussein A, Gaber MM, Elyan E, Jayne C (2017) Imitation learning: a survey of learning methods. ACM Comput Surv (CSUR) 50(2):1–35. https://doi.org/10.1145/3054912
Silver D, Huang A, Maddison CJ, Guez A, Sifre L, Van Den Driessche G, Schrittwieser J, Antonoglou I, Panneershelvam V, Lanctot M et al (2016) Mastering the game of go with deep neural networks and tree search. Nature 529(7587):484–489. https://doi.org/10.1038/nature16961
Silver D, Hubert T, Schrittwieser J, Antonoglou I, Lai M, Guez A, Lanctot M, Sifre L, Kumaran D, Graepel T et al (2018) A general reinforcement learning algorithm that masters chess, shogi, and go through self-play. Science 362(6419):1140–1144. https://doi.org/10.1126/science.aar6404
Haarnoja T, Zhou A, Abbeel P, Levine S (2018) Soft actor-critic: off-policy maximum entropy deep reinforcement learning with a stochastic actor. In: International conference on machine learning, pp 1861–1870. PMLR
Johannink T, Bahl S, Nair A, Luo J, Kumar A, Loskyll M, Ojea JA, Solowjow E, Levine S (2019) Residual reinforcement learning for robot control. In: 2019 international conference on robotics and automation (ICRA), pp 6023–6029. https://doi.org/10.1109/ICRA.2019.8794127. IEEE
Wang L, Zhang W, He X, Zha H (2018) Supervised reinforcement learning with recurrent neural network for dynamic treatment recommendation. In: Proceedings of the 24th ACM SIGKDD international conference on knowledge discovery & data mining, pp 2447–2456. https://doi.org/10.1145/3219819.3219961
Yu C, Liu J, Nemati S, Yin G (2021) Reinforcement learning in healthcare: a survey. ACM Comput Surv (CSUR) 55(1):1–36. https://doi.org/10.1145/3477600
Tassa Y, Doron Y, Muldal A, Erez T, Li Y, Casas DL, Budden D, Abdolmaleki A, Merel J, Lefrancq A et al (2018) Deepmind control suite. arXiv:1801.00690. https://doi.org/10.48550/arXiv.1801.00690
Prudencio RF, Maximo MR, Colombini EL (2023) A survey on offline reinforcement learning: taxonomy, review, and open problems. IEEE Trans Neural Netw Learn Syst. https://doi.org/10.1109/TNNLS.2023.3250269
Liu M, Zhao H, Yang Z, Shen J, Zhang W, Zhao L, Liu T-Y (2021) Curriculum offline imitating learning. Adv Neural Inf Process Syst 3:4
Kostrikov I, Nair A, Levine S (2021) Offline reinforcement learning with implicit q-learning. In: Deep RL workshop NeurIPS 2021
Xu H, Jiang L, Jianxiong L, Zhan X (2022) A policy-guided imitation approach for offline reinforcement learning. Adv Neural Inf Process Syst 35:4085–4098
Xu H, Jiang L, Li J, Yang Z, Wang Z, Chan VWK, Zhan X (2022) Offline rl with no ood actions: In-sample learning via implicit value regularization. In: The eleventh international conference on learning representations
Snell CV, Kostrikov I, Su Y, Yang S, Levine S (2022) Offline rl for natural language generation with implicit language q learning. In: The eleventh international conference on learning representations
Zheng Q, Henaff M, Amos B, Grover A (2023) Semi-supervised offline reinforcement learning with action-free trajectories. In: International conference on machine learning, pp 42339–42362. PMLR
Kumar A, Fu J, Soh M, Tucker G, Levine S (2019) Stabilizing off-policy q-learning via bootstrapping error reduction. In: Advances in neural information processing systems, vol 32
Liu Y, Swaminathan A, Agarwal A, Brunskill E (2020) Off-policy policy gradient with stationary distribution correction. In: Uncertainty in artificial intelligence, pp. 1180–1190. PMLR
Rashidinejad P, Zhu H, Yang K, Russell S, Jiao J (2022) Optimal conservative offline rl with general function approximation via augmented Lagrangian. In: The eleventh international conference on learning representations
Rafailov R, Yu T, Rajeswaran A, Finn C (2021) Offline reinforcement learning from images with latent space models. In: Learning for dynamics and control, pp 1154–1168. PMLR
Argenson A, Dulac-Arnold G (2020) Model-based offline planning. arXiv:2008.05556. https://doi.org/10.48550/arXiv.2008.05556
Hong Z-W, Agrawal P, Combes RT, Laroche R (2022) Harnessing mixed offline reinforcement learning datasets via trajectory weighting. In: The eleventh international conference on learning representations
Kumar A, Zhou A, Tucker G, Levine S (2020) Conservative q-learning for offline reinforcement learning. Adv Neural Inf Process Syst 33:1179–1191
Yu T, Kumar A, Rafailov R, Rajeswaran A, Levine S, Finn C (2021) Combo: conservative offline model-based policy optimization. Adv Neural Inf Process Syst 34:28954–28967
Shi L, Li G, Wei Y, Chen Y, Chi Y (2022) Pessimistic q-learning for offline reinforcement learning: towards optimal sample complexity. In: International conference on machine learning, pp 19967–20025. PMLR
Ha D, Schmidhuber J (2018) World models. arXiv:1803.10122. https://doi.org/10.48550/arXiv.1803.10122
Chen AS, Nam H, Nair S, Finn C (2021) Batch exploration with examples for scalable robotic reinforcement learning. IEEE Robot Autom Lett 6(3):4401–4408. https://doi.org/10.1109/LRA.2021.3068655
Driess D, Schubert I, Florence P, Li Y, Toussaint M (2022) Reinforcement learning with neural radiance fields. Adv Neural Inf Process Syst 35:16931–16945
Yi Q, Zhang R, Peng S, Guo J, Hu X, Du Z, Guo Q, Chen R, Li L, Chen Y (2023) Learning controllable elements oriented representations for reinforcement learning. Neurocomputing. https://doi.org/10.1016/j.neucom.2023.126455
Cui B, Chow Y, Ghavamzadeh M (2020) Control-aware representations for model-based reinforcement learning. arXiv:2006.13408. https://doi.org/10.48550/arXiv.2006.13408
Laskin M, Srinivas A, Abbeel P (2020) Curl: cunsupervised representations for reinforcement learning. In: International conference on machine learning, pp 5639–5650. PMLR
Ma G, Wang Z, Yuan Z, Wang X, Yuan B, Tao D (2022) A comprehensive survey of data augmentation in visual reinforcement learning. arXiv:2210.04561
Nair AV, Pong V, Dalal M, Bahl S, Lin S, Levine S (2018) Visual reinforcement learning with imagined goals. In: Advances in neural information processing systems, vol 31
Gelada C, Kumar S, Buckman J, Nachum O, Bellemare MG (2019) Deepmdp: learning continuous latent space models for representation learning. In: International conference on machine learning, pp 2170–2179. PMLR
Hafner D, Lillicrap T, Ba J, Norouzi M (2019) Dream to control: learning behaviors by latent imagination. In: International conference on learning representations
Hafez MB, Weber C, Kerzel M, Wermter S (2019) Efficient intrinsically motivated robotic grasping with learning-adaptive imagination in latent space. In: 2019 Joint IEEE 9th international conference on development and learning and epigenetic robotics (Icdl-Epirob). IEEE, pp 1–7. https://doi.org/10.1109/DEVLRN.2019.8850723
Hafez MB, Weber C, Kerzel M, Wermter S (2020) Improving robot dual-system motor learning with intrinsically motivated meta-control and latent-space experience imagination. Robot Auton Syst 133:103630. https://doi.org/10.1016/j.robot.2020.103630
Han D, Doya K, Tani J (2019) Variational recurrent models for solving partially observable control tasks. arXiv:1912.10703. https://doi.org/10.48550/arXiv.1912.10703
Lee AX, Nagabandi A, Abbeel P, Levine S (2020) Stochastic latent actor-critic: deep reinforcement learning with a latent variable model. Adv Neural Inf Process Syst 33:741–752
Garcıa J, Fernández F (2015) A comprehensive survey on safe reinforcement learning. J Mach Learn Res 16(1):1437–1480
Fei Y, Yang Z, Chen Y, Wang Z (2021) Exponential bellman equation and improved regret bounds for risk-sensitive reinforcement learning. In: Advances in neural information processing systems, vol 34
Zhang K, Zhang X, Hu B, Basar T (2021) Derivative-free policy optimization for linear risk-sensitive and robust control design: implicit regularization and sample complexity. In: Advances in neural information processing systems, vol 34
Greenberg I, Chow Y, Ghavamzadeh M, Mannor S (2022) Efficient risk-averse reinforcement learning. Adv Neural Inf Process Syst 35:32639–32652
Théate T, Ernst D (2023) Risk-sensitive policy with distributional reinforcement learning. Algorithms 16(7):325. https://doi.org/10.3390/a16070325
Rabin M (2000) Risk aversion and expected-utility theory: a calibration theorem. Econometrica 68(5):1281–1292
Tversky A, Kahneman D (1992) Advances in prospect theory: cumulative representation of uncertainty. J Risk Uncertain 5(4):297–323
Rockafellar RT, Uryasev S (2002) Conditional value-at-risk for general loss distributions. J Bank Finance 26(7):1443–1471
Sarykalin S, Serraino G, Uryasev S (2008) Value-at-risk vs. conditional value-at-risk in risk management and optimization. In: State-of-the-art decision-making tools in the information-intensive age. Informs, Maryland, pp 270–294. https://doi.org/10.1287/educ.1080.0052
Artzner P, Delbaen F, Eber J-M, Heath D (1999) Coherent measures of risk. Math Finance 9(3):203–228
Pinto L, Davidson J, Sukthankar R, Gupta A (2017) Robust adversarial reinforcement learning. In: International conference on machine learning. PMLR, pp 2817–2826
Chow Y, Ghavamzadeh M (2014) Algorithms for cvar optimization in mdps. In: Advances in neural information processing systems, vol 27
Chow Y, Tamar A, Mannor S, Pavone M (2015) Risk-sensitive and robust decision-making: a cvar optimization approach. In: Advances in neural information processing systems, vol 28
Ying C, Zhou X, Su H, Yan D, Chen N, Zhu J (2022) Towards safe reinforcement learning via constraining conditional value-at-risk. arXiv:2206.04436. https://doi.org/10.48550/arXiv.2206.04436
Ma X, Xia L, Zhou Z, Yang J, Zhao Q (2020) Dsac: distributional actor critic for risk-sensitive reinforcement learning. arXiv:2004.14547. https://doi.org/10.48550/arXiv.2004.14547
Armengol Urpí N, Curi S, Krause A (2021) Risk-averse offline reinforcement learning. In: International conference on learning representations (ICLR 2021). OpenReview
Fujimoto S, Meger D, Precup D (2019) Off-policy deep reinforcement learning without exploration. In: International conference on machine learning. PMLR, pp 2052–2062
Ma Y, Jayaraman D, Bastani O (2021) Conservative offline distributional reinforcement learning. In: Advances in neural information processing systems, vol 34
Rockafellar RT (2007) Coherent approaches to risk in optimization under uncertainty. In: OR tools and applications: glimpses of future technologies. Informs, Maryland, USA, pp 38–61
Delbaen F (2002) Coherent risk measures on general probability spaces. In: Advances in finance and stochastics. Springer, Berlin, pp 1–37. https://doi.org/10.1007/978-3-662-04790-3_1
Rockafellar RT, Uryasev SP, Zabarankin M (2002) Deviation measures in risk analysis and optimization. University of Florida, Department of Industrial & Systems Engineering Working Paper (2002–7)
Wang SS (2000) A class of distortion operators for pricing financial and insurance risks. J Risk Insur. https://doi.org/10.2307/253675
Ahmadi-Javid A (2011) An information-theoretic approach to constructing coherent risk measures. In: 2011 IEEE international symposium on information theory proceedings. IEEE, pp 2125–2127. https://doi.org/10.1109/ISIT.2011.6033932
Rockafellar RT, Uryasev S et al (2000) Optimization of conditional value-at-risk. J Risk 2:21–42
Rockafellar RT, Uryasev S, Zabarankin M (2006) Generalized deviations in risk analysis. Finance Stoch 10(1):51–74. https://doi.org/10.1007/s00780-005-0165-8
Levine S (2018) Reinforcement learning and control as probabilistic inference: Tutorial and review. arXiv:1805.00909. https://doi.org/10.48550/arXiv.1805.00909
Kim S, Jo H, Song J-B (2022) Object manipulation system based on image-based reinforcement learning. Intell Serv Robot. https://doi.org/10.1007/s11370-021-00402-6
Hafner D, Lillicrap T, Fischer I, Villegas R, Ha D, Lee H, Davidson J (2019) Learning latent dynamics for planning from pixels. In: International conference on machine learning. PMLR, pp 2555–2565
Odaibo S (2019) Tutorial: deriving the standard variational autoencoder (vae) loss function. arXiv:1907.08956. https://doi.org/10.48550/arXiv.1907.08956
Huber PJ (1964) Robust estimation of a location parameter. Ann Math Stat 35:73–101
Acerbi C (2002) Spectral measures of risk: a coherent representation of subjective risk aversion. J Bank Finance 26(7):1505–1518. https://doi.org/10.1016/S0378-4266(02)00281-9
Duan J, Guan Y, Li SE, Ren Y, Sun Q, Cheng B (2021) Distributional actor-critic: off-policy reinforcement learning for addressing value estimation errors. IEEE Trans Neural Netw Learn Syst. https://doi.org/10.1109/TNNLS.2021.3082568
Dabney W, Ostrovski G, Silver D, Munos R (2018) Implicit quantile networks for distributional reinforcement learning. In: International conference on machine learning. PMLR, pp 1096–1105
Ba JL, Kiros JR, Hinton GE (2016) Layer normalization. arXiv:1607.06450
Fujimoto S, Hoof H, Meger D (2018) Addressing function approximation error in actor-critic methods. In: International conference on machine learning. PMLR, pp 1587–1596
Fu J, Kumar A, Nachum O, Tucker G, Levine S (2020) D4rl: datasets for deep data-driven reinforcement learning. arXiv:2004.07219. https://doi.org/10.48550/arXiv.2004.07219
Agarwal R, Schwarzer M, Castro PS, Courville AC, Bellemare M (2021) Deep reinforcement learning at the edge of the statistical precipice. Adv Neural Inf Process Syst 34:29304–29320
Acknowledgements
We would like to thank the emergency medical services dispatch center from the Centre Hospitalier Universitaire Vaudois (CHUV) and more specifically Dr. Fabrice Dami for interesting discussions, that hat highlighted the necessity to develop high-dimensional risk-sensitive RL algorithms.
Funding
Open access funding provided by University of Applied Sciences and Arts Western Switzerland (HES-SO). This work was funded by Interreg France-Suisse (project SIA-REMU) (https://www.interreg-francesuisse.eu/).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors have no relevant financial or non-financial interest to disclose.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix A Experimental results
Appendix A Experimental results
In this section, we present our complete empirical results. More precisely, you can find all results of the deterministic environment in Table 3, and all results of the stochastic environment can be found in Table 4.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Hêche, F., Barakat, O., Desmettre, T. et al. Offline reinforcement learning in high-dimensional stochastic environments. Neural Comput & Applic 36, 585–598 (2024). https://doi.org/10.1007/s00521-023-09029-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-023-09029-3