张笃振;程翔;王岩松;张新贺;张铁林;杜久林;徐波;

• 1:中国科学院自动化研究所
• 2:中国科学院神经所

摘要(Abstract)：

类脑智能研究深入交叉脑科学和人工智能，旨在从脑科学中汲取结构、功能、机制等方面的灵感，用以启发人工智能软硬件研究。本文聚焦生物结构，重点总结神经侧向交互、生物彩票网络假设、Mot件架构的结构设计中。未来，随着多尺度和多类型生物网络组图谱的绘制，越来越多生物结构启发的网络基本算子可以被抽提出来并持续推动类脑智能的创新发展。

关键词(KeyWords)： 生物结构;网络算子;类脑智能;人工神经网络

Abstract:

Keywords:

基金项目(Foundation): 科技创新2030新一代人工智能项目（2020AAA0104305）;; 中科院先导专项A(XDA27010000)、先导专项B(XDB32070000);; 上海市市级科技重大专项（2021SHZDZX）

作者(Authors): 张笃振;程翔;王岩松;张新贺;张铁林;杜久林;徐波;

参考文献(References)：

• [1]COOK S J,JARRELL T A,BRITTIN C A,et al.Whole-animal connectomes of both Caenorhabditis elegans sexes[J].Nature,2019,571(7763):63-71.
• [2]KUNST M,LAURELL E,MOKAYES N,et al.A cellular-resolution atlas of the larval zebrafsh brain[J].Neuron,2019,103(1):21-38.
• [3]SEUNG WOOK OH,JULIE A HARRIS,LYDIA NG,et al.A mesoscale connectome of the mouse brain.Nature,2014.508(7495):207-214.
• [4]FIORENTINI A.Mach band phenomena[M].Visual psychophysics.Berlin:Springer,1972:188-201.
• [5]HARTLINE H K,WAGNER H G,RATLIFF F.Inhibition in the eye of Limulus[J].The Journal of general physiology,1956,39(5):651-673.
• [6]BLAKEMORE C,CARPENTER R H S,GEORGESON M A.Lateral inhibition between orientation detectors in the human visual system[J].Nature,1970,228(5266):37-39.
• [7]URBAN N N.Lateral inhibition in the olfactory bulb and in olfaction[J].Physiology&behavior,2002,77(4-5):607-612.
• [8]SINGER W,CREUTZFELDT O D.Reciprocal lateral inhibition of on-and off-center neurones in the lateral geniculate body of the cat[J].Experimental Brain Research,1970,10(3):311-330.
• [9]WANNER A A,GENOUD C,MASUDI T,et al.Dense EM-based reconstruction of the interglomerular projectome in the zebrafsh olfactory bulb[J].Nature neuroscience,2016,19(6):816-825.
• [10]FRIEDRICH R W,WANNER A A.Dense circuit reconstruction to understand neuronal computation:focus on zebrafsh[J].Annual Review of Neuroscience,2021,44:275-293.
• [11]CHRISTIE J M,WESTBROOK G L.Lateral excitation within the olfactory bulb[J].Journal of Neuroscience,2006,26(8):2269-2277.
• [12]NOWOTNY T,RABINOVICH M I,HUERTA R,et al.Decoding temporal information through slow lateral excitation in the olfactory system of insects[J].Journal of computational neuroscience,2003,15(2):271-281.
• [13]HINTON G E,SRIVASTAVA N,KRIZHEVSKY A,et al.Improving neural networks by preventing co-adaptation of feature detectors[EB/OL].(2012-07-03)[2022-11-30].https://arxiv.org/abs/1207.0580.
• [14]DIEHL P U,COOK M.Unsupervised learning of digit recognition using spike-timing-dependent plasticity[J].Frontiers in computational neuroscience,2015,9:99.
• [15]NGUYEN H T,NGUYEN T P,VU T N,et al.Outward influence and cascade size estimation in billion-scale networks[J].Proceedings of the ACM on Measurement and Analysis of Computing Systems,2017,1(1):1-30.
• [16]MARTEL J N P,SANDAMIRSKAYA Y.A neuromorphic approach for tracking using dynamic neural felds on a programmable vision-chip[C]//Proceedings of the 10th International Conference on Distributed Smart Camera.2016:148-154.
• [17]CHENG X,HAO Y,XU J,et al.LISNN:Improving spiking neural networks with lateral interactions for robust object recognition[C]//IJCAI.2020:1519-1525.
• [18]COX D D,DEAN T.Neural networks and neuroscience-inspired computer vision[J].Current Biology,2014,24(18):921-929.
• [19]FRANKLE J,CARBIN M.The Lottery Ticket Hypothesis:Finding Sparse,Trainable Neural Networks[C]//International Conference on Learning Representations.ICRL Organization,2018.
• [20]MALACH E,YEHUDAI G,SHALEV-SCHWARTZ S,et al.Proving the lottery ticket hypothesis:Pruning is all you need[C]//International Conference on Machine Learning.PMLR,2020:6682-6691.
• [21]FRANKLE J,DZIUGAITE G K,ROY D,et al.Linear mode connectivity and the lottery ticket hypothesis[C]//International Conference on Machine Learning.PMLR,2020:3259-3269.
• [22]ZHOU H,LAN J,LIU R,et al.Deconstructing lottery tickets:Zeros,signs,and the supermask[J].Advances in neural information processing systems,2019,32.
• [23]WORTSMAN M,RAMANUJAN V,LIU R,et al.Supermasks in superposition[J].Advances in Neural Information Processing Systems,2020,33:15173-15184.
• [24]WHITE J G,SOUTHGATE E,THOMSON J N,et al.The structure of the nervous system of the nematode Caenorhabditis elegans:the mind of a worm[J].Philosophical transactions of the Royal Society of London.Series B,Biological sciences,1986,314(1):340.
• [25]CHALFIE M,SULSTON J E,WHITE J G,et al.The neural circuit for touch sensitivity in Caenorhabditis elegans[J].Journal of Neuroscience,1985,5(4):956-964.
• [26]ISLAM M A,WANG Q,HASANI R M,et al.Probabilistic reachability analysis of the tap withdrawal circuit in caenorhabditis elegans[C]//2016 IEEE International High Level Design Validation and Test Workshop (HLDVT).IEEE,2016:170-177.
• [27]SUSOY V,HUNG W,WITVLIET D,et al.Natural sensory context drives diverse brain-wide activity during C.elegans mating[J].Cell,2021,184(20):5122-5137.
• [28]WITVLIET D,MULCAHY B,MITCHELL J K,et al.Connectomes across development reveal principles of brain maturation[J].Nature,2021,596(7871):257-261.
• [29]BRITTIN C A,COOK S J,HALL D H,et al.A multi-scale brain map derived from whole-brain volumetric reconstructions[J].Nature,2021,591(7848):105-110.
• [30]RANKIN C H,BECK C D O,CHIBA C M.Caenorhabditis elegans:a new model system for the study of learning and memory[J].Behavioural brain research,1990,37(1):89-92.
• [31]WICKS S R,ROEHRIG C J,RANKIN C H.A dynamic network simulation of the nematode tap withdrawal circuit:predictions concerning synaptic function using behavioral criteria[J].Journal of Neuroscience,1996,16(12):4017-4031.
• [32]SINGH S P,SUTTON R S.Reinforcement learning with replacing eligibility traces[J].Machine learning,1996,22(1):123-158.
• [33]HASANI R,LECHNER M,AMINI A,et al.A natural lottery ticket winner:Reinforcement learning with ordinary neural circuits[C]//International Conference on Machine Learning.PMLR,2020:4082-4093.
• [34]HASANI R M,LECHNER M,AMINI A,et al.Liquid time-constant recurrent neural networks as universal approximators[EB/OL].(2018-11-01)[2022-11-23].https://arxiv.org/abs/1811.00321.
• [35]LECHNER M,HASANI R,AMINI A,et al.Neural circuit policies enabling auditable autonomy[J].Nature Machine Intelligence,2020,2(10):642-652.
• [36]ZHANG D Z,ZHANG T L,JIA S C,et al.Multiscale Dynamic Coding improved Spiking Actor Network for Reinforcement Learning[C]//Thirty-Sixth AAAI Conference on Artifcial Intelligence (Virtual conference).AAAI,2022.
• [37]HAN X,JIA K,ZHANG T.Mouse-Brain Topology improved Evolutionary Neural Network for Efficient Reinforcement Learning[C]//International Conference on Intelligence Science (ICIS 2022).CAAI,2022.
• [38]ZHANG D,ZHANG T,JIA S,et al.Population-coding and dynamic-neurons improved spiking actor network for reinforcement learning[EB/OL].(2022-09-22)[2022-11-30].https://arxiv.org/abs/2106.07854.
• [39]HAGMANN P.From diffusion MRI to brain connectomics[R].EPFL,2005.
• [40]SPORNS O,TONONI G,K?TTER R.The human connectome:a structural description of the human brain[J].PLo Scomputational biology,2005,1(4):e42.
• [41]ZHONG Q,LI A,JIN R,et al.High-defnition imaging using line-illumination modulation microscopy[J].Nature methods,2021,18(3):309-315.
• [42]WANG Q,DING S L,LI Y,et al.The Allen mouse brain common coordinate framework:a 3D reference atlas[J].Cell,2020,181(4):936-953.
• [43]GLASSER M F,COALSON T,ROBINSON E,et al.A multi-modal parcellation of human cerebral cortex[J].Nature,2016,536(7615):171.
• [44]PARISI G I,KEMKER R,PART J L,et al.Continual lifelong learning with neural networks:A review[J].Neural Networks,2019,113:54-71.
• [45]BITZENHOFER S H,SIEBEN K,SIEBERT K D,et al.Oscillatory activity in developing prefrontal networks results from thetagamma-modulated synaptic inputs[J].Cell reports,2015,11(3):486-497.
• [46]ABBOTT L F,BOCK D D,CALLAWAY E M,et al.The mind of a mouse[J].Cell,2020,182(6):1372-1376.