Hardware architecture of the Protein Processing Associative Memory and the effects of dimensionality and quantisation on performance
Tóm tắt
The Protein Processor Associative Memory (PPAM) is a novel hardware architecture for a distributed, decentralised, robust and scalable, bidirectional, hetero-associative memory, that can adapt online to changes in the training data. The PPAM uses the location of data in memory to identify relationships and is therefore fundamentally different from traditional processing methods that tend to use arithmetic operations to perform computation. This paper presents the hardware architecture and details a sample digital logic implementation with an analysis of the implications of using existing techniques for such hardware architectures. It also presents the results of implementing the PPAM for a robotic application that involves learning the forward and inverse kinematics. The results show that, contrary to most other techniques, the PPAM benefits from higher dimensionality of data, and that quantisation intervals are crucial to the performance of the PPAM.
Tài liệu tham khảo
J. Amaral, J. Ghosh, An associative memory architecture for concurrent production systems. in Proceedings of the IEEE International Conference on Systems, Man, and Cybernetics, vol. 3 (1994), pp. 2219–2224. doi:10.1109/ICSMC.1994.400194.
G. Anzellotti, R. Battiti, I. Lazzizzera, G. Soncini, A. Zorat, A. Sartori, G. Tecchiolli, P. Lee, Totem: a highly parallel chip for triggering applications with inductive learning based on the reactive tabu search. Int. J. Mod. Phys. C 6(4), 555–560 (1995)
J. Backus, Can programming be liberated from the Von Neumann style? A functional style and its algebra of programs. Commun. ACM 21(8), 613–641 (1978). doi:10.1145/359576.359579
K. Chellapilla, D. Fogel, Evolving neural networks to play checkers without relying on expert knowledge. IEEE Trans. Neural Netw. 10(6), 1382–1391 (1999). doi:10.1109/72.809083
B. Coppin, Artificial Intelligence Illuminated, 1st edn. (Jones and Bartlett Publishers Inc, Sudbury, MA, 2004)
D.B. Fogel, Blondie24: Playing at the Edge of AI (Morgan Kaufmann Publishers Inc., San Francisco, CA, 2002)
A. Grübl, VLSI Implementation of a Spiking Neural Network. PhD thesis (University of Heidelberg, Heidelberg, 2007)
S. Hauck, Asynchronous design methodologies: an overview. Proc. IEEE 83(1), 69–93 (1995). doi:10.1109/5.362752
D.O. Hebb, The Organization of Behavior: A Neuropsychological Theory (Wiley, New York, 1949)
M. Hülse, S. McBride, M. Lee, Fast learning mapping schemes for robotic hand-eye coordination. Cogn. Comput. 2, 1–16 (2010). doi:10.1007/s12559-009-9030-y
Intel (2011) Intel 64 and IA-32 Architectures Optimization Reference Manual. Intel, 248966th edn
C.G. Johnson, The non-classical mind: cognitive science and non-classical computing. in Intelligent Computing Everywhere, chap 3, ed. by A. Schuster (Springer, Berlin, 2007), pp. 45–59. doi:10.1007/978-1-84628-943-9_3
B. Kosko, Bidirectional associative memories. IEEE Trans. Syst. Man Cybern. 18(1), 49–60 (1988). doi:10.1109/21.87054
S.W. Lee, J.T. Kim, H. Wang, D.J. Bae, K.M. Lee, J.H. Lee, J.W. Jeon, Architecture of RETE network hardware accelerator for real-time context-aware system. in KES (1), vol. 4251 ed. by B. Gabrys, R.J. Howlett, L.C. Jain (Springer, LNCS, Berlin, 2006), pp. 401–408
A. Lenz, S. Skachek, K. Hamann, J. Steinwender, A. Pipe, C. Melhuish, The bert2 infrastructure: an integrated system for the study of human-robot interaction. in IEEE-Humanoids-2010 (2010), pp. 346–351. doi:10.1109/ICHR.2010.5686319
T. Pfeil, A. Grübl, S. Jeltsch, E. Müller, P. Müller, M.A. Petrovici, M. Schmuker, D. Brüderle, J. Schemmel, K. Meier, Six networks on a universal neuromorphic computing substrate. Front. Neurosci. 7, 11 (2013)
O. Qadir, J. Liu, J. Timmis, G. Tempesti, A. Tyrrell, Principles of protein processing for a self-organising associative memory. in Proceedings of the IEEE CEC 2010 (2010)
O. Qadir, J. Liu, J. Timmis, G. Tempesti, A. Tyrrell, From bidirectional associative memory to a noise-tolerant, robust self-organising associative memory. Artif. Intell. 175(2), 673–693 (2011). doi:10.1016/j.artint.2010.10.008
O. Qadir, J. Liu, J. Timmis, G. Tempesti, A. Tyrrell, Hardware architecture for a bidirectional hetero-associative protein processing associative Memory. in Proceedings of the IEEE CEC 2011, (New Orleans, 2011)
O. Qadir, J. Timmis, G. Tempesti, A. Tyrrell, The protein processor associative memory on a robotic hand-eye coordination task. in 6th International ICST Conference on Bio-Inspired Models of Network (Information and Computing Systems (Bionetics2011), York, 2011)
O. Qadir, J. Timmis, G. Tempesti, A. Tyrrell, Profiling the fault tolerance for the adaptive protein processing associative memory. in NASA/ESA Conference on Adaptive Hardware and Systems (AHS2012) (Nuremberg, Germany, 2012)
C. Sakellariou, P.J. Bentley, Describing the fpga-based hardware architecture of systemic computation (haos). Comput. Inform. 31(3), 1001–1021 (2012)
M. Samie, G. Dragffy, A. Popescu, T. Pipe, C. Melhuish, Prokaryotic bio-inspired model for embryonics. in Proceedings of AHS 2009 (IEEE Computer Society, Washington, DC, AHS ’09, 2009), pp. 163–170. doi:10.1109/AHS.2009.45
A. Sudo, A. Sato, O. Hasegawa, Associative memory for online learning in noisy environments using self-organizing incremental neural. Network 20(6), 964–972 (2009). doi:10.1109/TNN.2009.2014374
J. Teich, Invasive algorithms and architectures (Invasive Algorithmen und Architekturen). IT—Inf. Technol. 50(5), 300–310 (2008)
K. Tirdad, A. Sadeghian, Hopfield neural networks as pseudo random number generators. in Fuzzy Information Processing Society (NAFIPS), 2010 Annual Meeting of the North American (2010), pp. 1–6. doi:10.1109/NAFIPS.2010.5548182
T. Toffoli, CAM: a high-performance cellular automaton machine. Phys. D 10, 195–204 (1984)
D. Tsafrir, Y. Etsion, D.G. Feitelson, S. Kirkpatrick, System noise, OS clock ticks, and fine-grained parallel applications. in Proceedings of the 19th Annual International Conference on Supercomputing (ACM, New York, NY, USA, ICS ’05, 2005), pp. 303–312. doi:10.1145/1088149.1088190
A. Turing, Computing machinery and intelligence. Mind 59, 433–460 (1950)
L. Wang, M. Jiang, R. Liu, X. Tang, Comparison bam and discrete hopfield networks with cpn for processing of noisy data. in Proceedings of ICSP2008 (2008), pp. 1708–1711. doi:10.1109/ICOSP.2008.4697466
C.C. Yang, S. Prasher, J.A. Landry, H. Ramaswamy, A. Ditommaso, Application of artificial neural networks in image recognition and classification of crop and weeds. Can. Agric. Eng. 42(3), 147–152 (2000)