Abstract
Unsupervised representation learning via generative modeling is a staple to many computer vision applications in the absence of labeled data. Variational Autoencoders (VAEs) are powerful generative models that learn representations useful for data generation. However, due to inherent challenges in the training objective, VAEs fail to learn useful representations amenable for downstream tasks. Regularization-based methods that attempt to improve the representation learning aspect of VAEs come at a price: poor sample generation. In this paper, we explore this representation-generation trade-off for regularized VAEs and introduce a new family of priors, namely decoupled priors, or dpVAEs, that decouple the representation space from the generation space. This decoupling enables the use of VAE regularizers on the representation space without impacting the distribution used for sample generation, and thereby reaping the representation learning benefits of the regularizations without sacrificing the sample generation. dpVAE leverages invertible networks to learn a bijective mapping from an arbitrarily complex representation distribution to a simple, tractable, generative distribution. Decoupled priors can be adapted to the state-of-the-art VAE regularizers without additional hyperparameter tuning. We showcase the use of dpVAEs with different regularizers. Experiments on MNIST, SVHN, and CelebA demonstrate, quantitatively and qualitatively, that dpVAE fixes sample generation and improves representation for regularized VAEs.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Complete derivation can be found in the supplementary material.
- 2.
The ELBOs for these regularizers can be found in the supplementary material.
- 3.
Architectures and hyperparameters are described in the supplementary material. Additionally, results showcasing that representation learning (specifically disentanglement) is not adversely affected by the introduction of decoupled priors are also presented in the supplementary material.
References
Zhao, S., Song, J., Ermon, S.: Infovae: balancing learning and inference in variational autoencoders. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 33, pp. 5885–5892 (2019)
Higgins, I., et al.: beta-vae: learning basic visual concepts with a constrained variational framework. ICLR 2, 6 (2017)
Kim, H., Mnih, A.: Disentangling by factorising. In: International Conference on Machine Learning, pp. 2654–2663 (2018)
Chen, X., Duan, Y., Houthooft, R., Schulman, J., Sutskever, I., Abbeel, P.: Infogan: interpretable representation learning by information maximizing generative adversarial nets. In: Advances in Neural Information Processing Systems, pp. 2172–2180 (2016)
Nguyen, A., Clune, J., Bengio, Y., Dosovitskiy, A., Yosinski, J.: Plug & play generative networks: conditional iterative generation of images in latent space. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4467–4477 (2017)
Mathieu, M.F., Zhao, J.J., Zhao, J., Ramesh, A., Sprechmann, P., LeCun, Y.: Disentangling factors of variation in deep representation using adversarial training. In: Advances in Neural Information Processing Systems, pp. 5040–5048 (2016)
Higgins, I., et al.: Darla: Improving zero-shot transfer in reinforcement learning. In: Proceedings of the 34th International Conference on Machine Learning-Volume 70, JMLR. org, pp. 1480–1490 (2017)
Rezende, D., Danihelka, I., Gregor, K., Wierstra, D., et al.: One-shot generalization in deep generative models. In: International Conference on Machine Learning, pp. 1521–1529 (2016)
Kingma, D.P., Mohamed, S., Rezende, D.J., Welling, M.: Semi-supervised learning with deep generative models. In: Advances in Neural Information Processing Systems, pp. 3581–3589 (2014)
Bengio, Y., Courville, A., Vincent, P.: Representation learning: a review and new perspectives. IEEE Trans. Pattern Anal. Mach. Intell. 35, 1798–1828 (2013)
Alemi, A., Poole, B., Fischer, I., Dillon, J., Saurous, R.A., Murphy, K.: Fixing a broken elbo. In: International Conference on Machine Learning, pp. 159–168 (2018)
Yuille, A., Kersten, D.: Vision as Bayesian inference: analysis by synthesis? Trends Cogn. Sci. 10, 301–308 (2006)
Nair, V., Susskind, J., Hinton, G.E.: Analysis-by-synthesis by learning to invert generative black boxes. In: Kůrková, V., Neruda, R., Koutník, J. (eds.) ICANN 2008. LNCS, vol. 5163, pp. 971–981. Springer, Heidelberg (2008). https://doi.org/10.1007/978-3-540-87536-9_99
Kingma, D.P., Welling, M.: Auto-encoding variational bayes. ICLR (2014)
Rezende, D.J., Mohamed, S., Wierstra, D.: Stochastic backpropagation and approximate inference in deep generative models. In: International Conference on Machine Learning, pp. 1278–1286 (2014)
Maaløe, L., Sønderby, C.K., Sønderby, S.K., Winther, O.: Auxiliary deep generative models. In: International Conference on Machine Learning, pp. 1445–1453 (2016)
Sønderby, C.K., Raiko, T., Maaløe, L., Sønderby, S.K., Winther, O.: How to train deep variational autoencoders and probabilistic ladder networks. In: 33rd International Conference on Machine Learning (ICML 2016) (2016)
Pu, Y., et al.: Variational autoencoder for deep learning of images, labels and captions. In: Advances in Neural Information Processing Systems, pp. 2352–2360 (2016)
Xu, W., Sun, H., Deng, C., Tan, Y.: Variational autoencoder for semi-supervised text classification. In: Thirty-First AAAI Conference on Artificial Intelligence (2017)
Tschannen, M., Bachem, O., Lucic, M.: Recent advances in autoencoder-based representation learning. In: Third workshop on Bayesian Deep Learning (NeurIPS 2018) (2018)
Chen, X., et al.: Variational lossy autoencoder. ICLR (2017)
Hoffman, M.D., Johnson, M.J.: Elbo surgery: yet another way to carve up the variational evidence lower bound. In: Workshop in Advances in Approximate Bayesian Inference, NIPS, vol. 1. (2016)
Chen, T.Q., Li, X., Grosse, R.B., Duvenaud, D.K.: Isolating sources of disentanglement in variational autoencoders. In: Advances in Neural Information Processing Systems, pp. 2610–2620 (2018)
Kumar, A., Sattigeri, P., Balakrishnan, A.: Variational inference of disentangled latent concepts from unlabeled observations. In: ICLR (2018)
Makhzani, A., Frey, B.J.: Pixelgan autoencoders. In: Advances in Neural Information Processing Systems, pp. 1975–1985 (2017)
Alemi, A.A., Fischer, I., Dillon, J.V., Murphy, K.: Deep variational information bottleneck. In: ICLR (2016)
Rosca, M., Lakshminarayanan, B., Mohamed, S.: Distribution matching in variational inference. arXiv preprint arXiv:1802.06847 (2018)
Xu, H., Chen, W., Lai, J., Li, Z., Zhao, Y., Pei, D.: On the necessity and effectiveness of learning the prior of variational auto-encoder. arXiv preprint arXiv:1905.13452 (2019)
Shmelkov, K., Lucas, T., Alahari, K., Schmid, C., Verbeek, J.: Coverage and quality driven training of generative image models. arXiv preprint arXiv:1901.01091 (2019)
Goodfellow, I., et al.: Generative adversarial nets. In: Advances in Neural Information Processing Systems, pp. 2672–2680 (2014)
Arjovsky, M., Chintala, S., Bottou, L.: Wasserstein generative adversarial networks. In: International conference on machine learning, pp. 214–223 (2017)
Sugiyama, M., Suzuki, T., Kanamori, T.: Density-ratio matching under the Bregman divergence: a unified framework of density-ratio estimation. Ann. Inst. Stat. Math. 64, 1009–1044 (2012)
Mescheder, L., Nowozin, S., Geiger, A.: Adversarial variational bayes: Unifying variational autoencoders and generative adversarial networks. In: Proceedings of the 34th International Conference on Machine Learning-Volume 70, JMLR. org, pp. 2391–2400 (2017)
Makhzani, A., Shlens, J., Jaitly, N., Goodfellow, I., Frey, B.: Adversarial autoencoders. arXiv preprint arXiv:1511.05644 (2015)
Srivastava, A., Valkov, L., Russell, C., Gutmann, M.U., Sutton, C.: Veegan: Reducing mode collapse in gans using implicit variational learning. In: Advances in Neural Information Processing Systems, pp. 3308–3318 (2017)
Makhzani, A., Shlens, J., Jaitly, N., Goodfellow, I., Frey, B.: Adversarial autoencoders. In: International Conference on Learning Representations (2016)
Tolstikhin, I., Bousquet, O., Gelly, S., Schoelkopf, B.: Wasserstein auto-encoders. arXiv preprint arXiv:1711.01558 (2017)
Xiao, Z., Yan, Q., Amit, Y.: Generative latent flow. arXiv preprint arXiv:1905.10485 (2019)
Kingma, D.P., Salimans, T., Jozefowicz, R., Chen, X., Sutskever, I., Welling, M.: Improved variational inference with inverse autoregressive flow. In: Advances in neural information processing systems, pp. 4743–4751(2016)
Bauer, M., Mnih, A.: Resampled priors for variational autoencoders. In: The 22nd International Conference on Artificial Intelligence and Statistics, pp. 66–75 (2019)
Tomczak, J., Welling, M.: Vae with a vampprior. In: International Conference on Artificial Intelligence and Statistics, pp. 1214–1223 (2018)
Dilokthanakul, N., et al.: Deep unsupervised clustering with Gaussian mixture variational autoencoders. arXiv preprint arXiv:1611.02648 (2016)
Gregor, K., Danihelka, I., Graves, A., Rezende, D., Wierstra, D.: Draw: a recurrent neural network for image generation. In: International Conference on Machine Learning, pp. 1462–1471(2015)
Gulrajani, I., et al.: Pixelvae: a latent variable model for natural images. In: ICLR (2017)
Van Den Oord, A., Vinyals, O., et al.: Neural discrete representation learning. In: Advances in Neural Information Processing Systems, pp. 6306–6315 (2017)
Razavi, A., van den Oord, A., Vinyals, O.: Generating diverse high-fidelity images with vq-vae-2. In: Advances in Neural Information Processing Systems, pp. 14866–14876 (2019)
Van den Oord, A., Kalchbrenner, N., Espeholt, L., Vinyals, O., Graves, A., et al.: Conditional image generation with pixelcnn decoders. In: Advances in Neural Information Processing Systems, pp. 4790–4798 (2016)
Oord, A.v.d., Kalchbrenner, N., Kavukcuoglu, K.: Pixel recurrent neural networks. arXiv preprint arXiv:1601.06759 (2016)
Dinh, L., Krueger, D., Bengio, Y.: Nice: Non-linear independent components estimation (2014)
Dinh, L., Sohl-Dickstein, J., Bengio, S.: Density estimation using real nvp. In: ICLR (2017)
Rezende, D., Mohamed, S.: Variational inference with normalizing flows. In: Proceedings of the 32nd International Conference on Machine Learning. Volume 37 of Proceedings of Machine Learning Research., Lille, France, PMLR, pp. 1530–1538 (2015)
Kingma, D.P., Dhariwal, P.: Glow: Generative flow with invertible 1 x 1 convolutions. In: Bengio, S., Wallach, H., Larochelle, H., Grauman, K., Cesa-Bianchi, N., Garnett, R. (eds.) Advances in Neural Information Processing Systems 31, pp. 10215–10224. Curran Associates, Inc. (2018)
Huang, C.W., et al.: Learnable explicit density for continuous latent space and variational inference. arXiv preprint arXiv:1710.02248 (2017)
Das, H.P., Abbeel, P., Spanos, C.J.: Dimensionality reduction flows. arXiv preprint arXiv:1908.01686 (2019)
Gritsenko, A.A., Snoek, J., Salimans, T.: On the relationship between normalising flows and variational-and denoising autoencoders (2019)
Bowman, S.R., Vilnis, L., Vinyals, O., Dai, A., Jozefowicz, R., Bengio, S.: Generating sentences from a continuous space. In: Proceedings of The 20th SIGNLL Conference on Computational Natural Language Learning, pp. 10–21 (2016)
Burgess, C.P., et al.: Understanding disentangling in beta-vae. arXiv preprint arXiv:1804.03599 (2018)
Liu, Q., Wang, D.: Stein variational gradient descent: A general purpose Bayesian inference algorithm. In: Advances in Neural Information Processing Systems, pp. 2378–2386 (2016)
Gretton, A., Borgwardt, K., Rasch, M., Schölkopf, B., Smola, A.J.: A kernel method for the two-sample-problem. In: Advances in Neural Information Processing Systems, pp. 513–520 (2007)
Li, Y., Swersky, K., Zemel, R.: Generative moment matching networks. In: International Conference on Machine Learning, pp. 1718–1727 (2015)
Dziugaite, G.K., Roy, D.M., Ghahramani, Z.: Training generative neural networks via maximum mean discrepancy optimization. In: Proceedings of the Thirty-First Conference on Uncertainty in Artificial Intelligence, pp. 258–267. AUAI Press (2015)
LeCun, Y., Cortes, C.: MNIST handwritten digit database (2010)
Netzer, Y., Wang, T., Coates, A., Bissacco, A., Wu, B., Ng, A.Y.: Reading digits in natural images with unsupervised feature learning. In: NIPS Workshop on Deep Learning and Unsupervised Feature Learning (2011)
Liu, Z., Luo, P., Wang, X., Tang, X.: Deep learning face attributes in the wild. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 3730–3738 (2015)
Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., Wojna, Z.: Rethinking the inception architecture for computer vision. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (2016)
Dinh, L., Sohl-Dickstein, J., Pascanu, R., Larochelle, H.: A RAD approach to deep mixture models. CoRR abs/1903.07714 (2019)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Bhalodia, R., Lee, I., Elhabian, S. (2021). dpVAEs: Fixing Sample Generation for Regularized VAEs. In: Ishikawa, H., Liu, CL., Pajdla, T., Shi, J. (eds) Computer Vision – ACCV 2020. ACCV 2020. Lecture Notes in Computer Science(), vol 12625. Springer, Cham. https://doi.org/10.1007/978-3-030-69538-5_39
Download citation
DOI: https://doi.org/10.1007/978-3-030-69538-5_39
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-69537-8
Online ISBN: 978-3-030-69538-5
eBook Packages: Computer ScienceComputer Science (R0)