Abstract
Spiking neural networks (SNNs) are well suited for functions such as data/pattern classification, estimation, prediction, signal processing and robotic control applications. Whereas the real-world embedded applications are often multi-functional with orthogonal or contradicting functional requirements. The EMBRACE hardware modular SNN architecture has been previously reported as an embedded computing platform for complex real-world applications. The EMBRACE architecture employs genetic algorithm (GA) for training the SNN which offers faster prototyping of SNN applications, but exhibits a number of limitations including poor scalability and search space explosions for the evolution of large-scale, complex, real-world applications. This paper investigates the limitations of evolving real-world embedded applications with orthogonal functional goals on hardware SNN using GA-based training. This paper presents a novel, fast and efficient application prototyping technique using the EMBRACE hardware modular SNN architecture and the GA-based evolution platform. Modular design and evolution of a robotic navigational controller application decomposed into obstacle avoidance controller and speed and direction manager application subtasks is presented. The proposed modular evolution technique successfully integrates the orthogonal functionalities of the application and helps to overcome contradicting application scenarios gracefully. Results illustrate that the modular evolution of the application reduces the SNN configuration search space and complexity for the GA-based SNN evolution, offering rapid and successful prototyping of complex applications on the hardware SNN platform. The paper presents validation results of the evolved robotic application implemented on the EMBRACE architecture prototyped on Xilinx Virtex-6 FPGA interacting with the player-stage robotics simulator.
Similar content being viewed by others
Notes
EMulating Biologically-inspiRed ArChitectures in hardwarE.
The task decomposition for the robotic navigational controller modular application design presented in this paper has been done manually based on input vector partitioning and analysing the orthogonality of the functional requirements of the application.
The markers for the proposed robotic navigational controller application are chosen by the designer such that the robot can progress towards the destination by following the markers in sequence. The markers can also be chosen by applying meta-heuristic algorithms to the classical Travelling Salesman Problem which is currently out of scope of this paper.
References
Haykin S (1998) Neural networks: a comprehensive foundation, 2nd edn. Prentice Hall PTR, Upper Saddle River
Maass W (1997) Networks of spiking neurons: the third generation of neural network models. Neural Netw 10(9):1659–1671
Gerstner W, Kistler WM (2002) Spiking neuron models: single neurons, populations, plasticity. Cambridge University Press, Cambridge
Bohte S, Kok JN (2005) Applications of spiking neural networks. Inf Process Lett 95(6):519–520
Pearson M, Pipe A, Mitchinson B, Gurney K, Melhuish C, Gilhespy I, Nibouche M (2007) Implementing spiking neural networks for real-time signal-processing and control applications: a model-validated FPGA approach. IEEE Trans Neural Netw 18(5):1472–1487
Harkin J, Morgan F, McDaid L, Hall S, McGinley B, Cawley S (2009) A reconfigurable and biologically inspired paradigm for computation using network-on-chip and spiking neural networks. Int J Reconfig Comput 2009:2:1–2:13
Cawley S, Morgan F, McGinley B, Pande S, McDaid L, Carrillo S, Harkin J (2011) Hardware spiking neural network prototyping and application. Genet Progr Evolvable Mach 12:257–280
Pande S, Morgan F, Cawley S, Bruintjes T, Smit G, McGinley B, Carrillo S, Harkin J, McDaid L (2013). Modular neural tile architecture for compact embedded hardware spiking neural network. Neural Process Lett 38(2):131–153
Pande S, Morgan F, Smit G, Bruintjes T, Rutgers J, McGinley B, Cawley S, Harkin J, McDaid L (2013) Fixed latency on-chip interconnect for hardware spiking neural network architectures. Parallel Comput 39:357–371
Pande S, Morgan F, Cawley S, McGinley B, Harkin J, Carrillo S, McDaid L (2011) Addressing the hardware resource requirements of network-on-chip based neural architectures. In: International conference on neural computation theory and applications (NCTA), Paris, France
Morgan F, Cawley S, McGinley B, Pande S, McDaid LJ, Glackin B, Maher J, Harkin J (2009) Exploring the evolution of NoC-based spiking neural networks on FPGAs. In: International conference on field-programmable technology, 2009 (FPT 2009). IEEE, pp 300–303
Brooks R (1986) A robust layered control system for a mobile robot. IEEE J Robot Autom 2(1):14–23
Vaughan R (2008) Massively multi-robot simulation in stage. Swarm Intell 2:189–208
Johannes S, Wieringa BM, Matzke M, Mnte TF (1996) Hierarchical visual stimuli: electrophysiological evidence for separate left hemispheric global and local processing mechanisms in humans. Neurosci Lett 210(2):111–114
Van Essen D, Anderson C, Felleman D (1992) Information processing in the primate visual system: an integrated systems perspective. Science 255(5043):419–423
Binzegger T, Douglas R, Martin K (2007) Stereotypical bouton clustering of individual neurons in cat primary visual cortex. J Neurosci 27(45):12242–12254
Happel B, Murre J (1994) Design and evolution of modular neural network architectures. Neural Netw 7(6):985–1004
Auda G, Kamel M (1999) Modular neural networks: a survey. Int J Neural Syst 9(02):129–151
Ronco E, Gawthrop P (1995) Modular neural networks: a state of the art. Rapport Technique CSC95026, Center of System and Control University of Glasgow 1:1–22
Osherson D, Weinstein S, Stob M (1993) Modular learning. MIT Press, Cambridge
Guan S, Li S, Tan SK (2004) Neural network task decomposition based on output partitioning. J Inst Eng Singap 44:78–89
Lu B-L, Ito M (1999) Task decomposition and module combination based on class relations: a modular neural network for pattern classification. IEEE Trans Neural Netw 10(5):1244–1256
Thangavelautham J, Deleuterio G (2004) A neuroevolutionary approach to emergent task decomposition. In: Proceedings of 8th parallel problem solving from nature. Springer, Berlin, pp 991–1000
Khare V, Yao X, Sendhoff B, Jin Y, Wersing H (2005) Co-evolutionary modular neural networks for automatic problem decomposition. In: The 2005 IEEE congress on evolutionary computation, 2005, vol 3. IEEE, pp 2691–2698
Santos J, Alexandre L, de Sá JM (2006) Modular neural network task decomposition via entropic clustering. In: Sixth international conference on intelligent systems design and applications, 2006. ISDA ’06, vol 1. IEEE, pp 62–67
Hebb DO (2005) The organization of behavior: a neuropsychological theory. Psychology Press, Hove
Hinton G, Sejnowski T (1999) Unsupervised learning: foundations of neural computation. The MIT press, Cambridge
Natschläger T, Ruf B (1998) Online clustering with spiking neurons using temporal coding. Prog Neural Process 10:33–42
Bohte S, La Poutré H, Kok J (2002) Unsupervised clustering with spiking neurons by sparse temporal coding and multilayer RBF networks. IEEE Trans Neural Netw 13(2):426–435
Landis F, Ott T, Stoop R (2010) Hebbian self-organizing integrate-and-fire networks for data clustering. Neural Comput 22(1):273–288
Hopfield JJ (1995) Pattern recognition computation using action potential timing for stimulus representation. Nature 376(6535):33–36
Gerstner W, van Hemmen JL (1992) Associative memory in a network of spiking neurons. Netw Comput Neural Syst 3(2):139–164
Zamani M, Sadeghian A, Chartier S (2010) A bidirectional associative memory based on cortical spiking neurons using temporal coding. In: IJCNN, pp 1–8
Alpaydin E (2010) Introduction to machine learning. The MIT Press, Cambridge
Bohte S, Kok J, La Poutre H (2002) Error-backpropagation in temporally encoded networks of spiking neurons. Neurocomputing 48(1):17–37
Legenstein R, Naeger C, Maass W (2005) What can a neuron learn with spike-timing-dependent plasticity? Neural Comput 17(11):2337–2382
Legenstein R, Pecevski D, Maass W (2008) A learning theory for reward-modulated spike-timing-dependent plasticity with application to biofeedback. PLoS Comput Biol 4(10):e1000180
Baras D, Meir R (2007) Reinforcement learning, spike-time-dependent plasticity, and the BCM rule. Neural Comput 19(8):2245–2279
Farries M, Fairhall A (2007) Reinforcement learning with modulated spike timing-dependent synaptic plasticity. J Neurophysiol 98(6):3648–3665
Florian R (2005) A reinforcement learning algorithm for spiking neural networks. In: Seventh international symposium on symbolic and numeric algorithms for scientific computing, 2005 (SYNASC 2005). IEEE
Florian R (2007) Reinforcement learning through modulation of spike-timing-dependent synaptic plasticity. Neural Comput 19(6):1468–1502
Izhikevich E (2007) Solving the distal reward problem through linkage of STDP and dopamine signaling. Cereb Cortex 17(10):2443–2452
Vasilaki E, Frémaux N, Urbanczik R, Senn W, Gerstner W (2009) Spike-based reinforcement learning in continuous state and action space: when policy gradient methods fail. PLoS Comput Biol 5(12):e1000586
Di Paolo E (2002) Spike-timing dependent plasticity for evolved robots. Adapt Behav 10(3–4):243–263
Hagras H, Pounds-Cornish A, Colley M, Callaghan V, Clarke G (2004) Evolving spiking neural network controllers for autonomous robots. In: 2004 IEEE international conference on robotics and automation, 2004. Proceedings. ICRA ’04 vol 5. IEEE, pp 4620–4626
Mitchell M (1998) An introduction to genetic algorithms. MIT Press, Cambridge
Vainbrand D, Ginosar R (2011) Scalable network-on-chip architecture for configurable neural networks. Microprocess Microsyst 35(2):152–166
Acknowledgments
This research is supported by the International Centre for Graduate Education in Micro and Nano-Engineering (ICGEE), Irish Research Council for Science, Engineering and Technology (IRCSET) and Xilinx University Programme.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Pande, S., Morgan, F., Krewer, F. et al. Rapid application prototyping for hardware modular spiking neural network architectures. Neural Comput & Applic 28, 2767–2779 (2017). https://doi.org/10.1007/s00521-015-2136-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-015-2136-0