雷鋒網(wǎng)AI科技評(píng)論按：當(dāng)前計(jì)算機(jī)技術(shù)面臨著兩個(gè)重要瓶頸：（1）摩爾定律失效；（2）「馮諾依曼」架構(gòu)導(dǎo)致的能效低下。

隨著集成電路的規(guī)模越來(lái)越接近物理極限，人類(lèi)若想進(jìn)一步提升計(jì)算機(jī)的性能，必然要考慮新的計(jì)算機(jī)架構(gòu)。而另一方面，「馮諾依曼」架構(gòu)中，運(yùn)算單元和存儲(chǔ)單元分離，使得大部分能量和時(shí)間都消耗在數(shù)據(jù)的讀取和存儲(chǔ)過(guò)程中；并且數(shù)據(jù)處理是基于串行結(jié)構(gòu)，即同一時(shí)刻只能執(zhí)行一個(gè)任務(wù)。

這與人腦處理信息的方式差別巨大。在進(jìn)行學(xué)習(xí)和認(rèn)知等復(fù)雜計(jì)算時(shí)，人腦的功耗只有 20 瓦；而目前最先進(jìn)的計(jì)算機(jī)模擬人腦功能，功耗也高達(dá) 800 萬(wàn)瓦以上，速度比人腦要慢 1000 倍以上。究其原因，是因?yàn)楝F(xiàn)代計(jì)算機(jī)一般使用固定的數(shù)字化的程序模型，同步、串行、集中、快速、具有通用性地處理問(wèn)題，數(shù)據(jù)存儲(chǔ)與計(jì)算過(guò)程在不同地址空間完成。而與之形成鮮明對(duì)比的是，人的大腦會(huì)重復(fù)利用神經(jīng)元，并突觸、異步、并行、分布式、緩慢、不具通用性地處理問(wèn)題，是可重構(gòu)的、專(zhuān)門(mén)的、容錯(cuò)的生物基質(zhì)，并且人腦記憶數(shù)據(jù)與進(jìn)行計(jì)算的邊界是模糊的。

鑒于此，當(dāng)前借鑒人腦發(fā)展的類(lèi)腦計(jì)算技術(shù)，被認(rèn)為是應(yīng)對(duì)當(dāng)前挑戰(zhàn)的重要方案?！邦?lèi)腦計(jì)算”本質(zhì)來(lái)說(shuō)，即利用神經(jīng)計(jì)算來(lái)模擬人類(lèi)大腦處理信息的過(guò)程，被認(rèn)為“下一代人工智能”的重要方向，也是當(dāng)前人工智能領(lǐng)域的熱點(diǎn)方向。

與經(jīng)典人工智能符號(hào)主義、連接主義、行為主義以及機(jī)器學(xué)習(xí)的統(tǒng)計(jì)主義這些技術(shù)路線不同，類(lèi)腦計(jì)算采取的是仿真主義：

• 結(jié)構(gòu)層次模仿腦（非馮·諾依曼體系結(jié)構(gòu)）

• 器件層次逼近腦（神經(jīng)形態(tài)器件替代晶體管）

• 智能層次超越腦（主要靠自主學(xué)習(xí)訓(xùn)練而不是人工編程）

因此，類(lèi)腦計(jì)算試圖建立一個(gè)類(lèi)似的架構(gòu)，使得計(jì)算機(jī)也能保持類(lèi)腦的復(fù)雜性，達(dá)到可處理小數(shù)據(jù) & 小標(biāo)注問(wèn)題、適用于弱監(jiān)督和無(wú)監(jiān)督問(wèn)題、關(guān)聯(lián)分析能力強(qiáng)、魯棒性強(qiáng)、計(jì)算資源消耗較少、具備認(rèn)知推理能力、時(shí)序相關(guān)性好、可能解決通用場(chǎng)景問(wèn)題的目的，最終實(shí)現(xiàn)強(qiáng)人工智能和通用智能。

目前，國(guó)際上類(lèi)腦計(jì)算研究已經(jīng)取得顯著進(jìn)展，技術(shù)探索階段已經(jīng)過(guò)去，技術(shù)預(yù)研已經(jīng)開(kāi)始，一些關(guān)鍵技術(shù)獲得突破，相關(guān)的技術(shù)原型和系統(tǒng)原型已開(kāi)發(fā)成功。

總的來(lái)說(shuō)，類(lèi)腦智能技術(shù)體系分四層：基礎(chǔ)理論層、硬件層、軟件層、產(chǎn)品層。

基礎(chǔ)理論層基于腦認(rèn)知與神經(jīng)計(jì)算，主要從生物醫(yī)學(xué)角度研究大腦可塑性機(jī)制、腦功能結(jié)構(gòu)、腦圖譜等大腦信息處理機(jī)制研究；

硬件層主要是實(shí)現(xiàn)類(lèi)腦功能的神經(jīng)形態(tài)芯片，也就是非馮諾依曼架構(gòu)的類(lèi)腦芯片，如脈沖神經(jīng)網(wǎng)絡(luò)芯片、憶阻器、憶容器、憶感器等；

軟件層包含核心算法和通用技術(shù)，核心算法主要是弱監(jiān)督學(xué)習(xí)和無(wú)監(jiān)督學(xué)習(xí)機(jī)器學(xué)習(xí)機(jī)制，如脈沖神經(jīng)網(wǎng)絡(luò)、增強(qiáng)學(xué)習(xí)、對(duì)抗神經(jīng)網(wǎng)絡(luò)等；

通用技術(shù)主要是包含視覺(jué)感知、聽(tīng)覺(jué)感知、多模態(tài)融合感知、自然語(yǔ)言理解、推理決策等；產(chǎn)品層主要包含交互產(chǎn)品和整機(jī)產(chǎn)品，交互產(chǎn)品包含腦機(jī)接口、腦控設(shè)備、神經(jīng)接口、智能假體等，整機(jī)產(chǎn)品主要有類(lèi)腦計(jì)算機(jī)、類(lèi)腦機(jī)器人等。

類(lèi)腦計(jì)算，被視為未來(lái)信息技術(shù)最具有發(fā)展前景的重要領(lǐng)域之一，正如歐盟人腦旗艦研究計(jì)劃所指出的：「在未來(lái) 20 到 30 年內(nèi)，誰(shuí)要想主導(dǎo)世界經(jīng)濟(jì)，誰(shuí)必須在類(lèi)腦計(jì)算這個(gè)領(lǐng)域領(lǐng)先」。

中國(guó)學(xué)術(shù)領(lǐng)域近年來(lái)也越來(lái)越關(guān)注類(lèi)腦計(jì)算的研究。這表現(xiàn)在多個(gè)方面。其一，國(guó)家層面上，從2016年起制定了為期15年的「腦計(jì)劃」，類(lèi)腦計(jì)算正是其核心研究領(lǐng)域之一。另一方面，在近些年來(lái)，有越來(lái)越多的會(huì)議開(kāi)始設(shè)置「類(lèi)腦計(jì)算」專(zhuān)場(chǎng)或論壇。例如前不久剛結(jié)束的由中國(guó)計(jì)算機(jī)學(xué)會(huì)、雷鋒網(wǎng)、香港中文大學(xué)共同舉辦的 CCF-GAIR 2019大會(huì)便設(shè)置了此專(zhuān)場(chǎng)。

近期將舉辦的國(guó)際圖像圖形學(xué)學(xué)術(shù)會(huì)議也將舉辦「類(lèi)腦智能論壇」。

國(guó)際圖象圖形學(xué)學(xué)術(shù)會(huì)議（ICIG）是中國(guó)圖象圖形學(xué)學(xué)會(huì)主辦的最高級(jí)別的系列國(guó)際會(huì)議，創(chuàng)建于2000年，每?jī)赡昱e辦一屆，迄今已經(jīng)成功舉辦九屆。

第十屆國(guó)際圖象圖形學(xué)學(xué)術(shù)會(huì)議（ICIG2019）將于2019年8月23-25日在北京友誼賓館召開(kāi)，主題為“人工智能時(shí)代的圖像圖形前沿研究”，由清華大學(xué)、北京大學(xué)和中國(guó)科學(xué)院自動(dòng)化研究所承辦，得到了國(guó)際模式識(shí)別協(xié)會(huì)（IAPR）的支持。本次的會(huì)議共包含了3個(gè)特邀報(bào)告、2個(gè)講習(xí)班、3個(gè)workshops，多個(gè)論壇。其中之一便為「類(lèi)腦智能論壇」。

類(lèi)腦智能論壇由CSIG機(jī)器視覺(jué)專(zhuān)委會(huì)承辦，由中科院自動(dòng)化所何暉光研究員和深圳職業(yè)技術(shù)學(xué)院人工智能學(xué)院院長(zhǎng)楊金峰教授共同組織，邀請(qǐng)到6位專(zhuān)家從不同的角度來(lái)介紹類(lèi)腦智能的研究進(jìn)展、類(lèi)腦研究中的難點(diǎn)問(wèn)題，并對(duì)今后的研究進(jìn)行展望。

各專(zhuān)家報(bào)告內(nèi)容可參考如下：

 

Si Wu

Peking University

Title：Push-pull Feedback Implements Rough-to-fine Information Processing

Abstract: Experimental data has revealed that in addition to feedforward connections, there exist abundant feedback connections in a hierarchical neural pathway. Although the importance of feedback in neural information processing has been widely recognized in the field, the detailed mechanism of how it works remains largely unknown. Here, we investigate the role of feedback in hierarchical memory retrieval. Specifically, we consider a multi-layer network which stores hierarchical memory patterns, and each layer of the network behaves as an associative memory of the corresponding hierarchy. We find that to achieve good retrieval performance, the feedback needs to be dynamical: at the early phase, the feedback is positive (push), which suppresses inter-class noises between memory patterns; at the late phase, the feedback is negative (pull), which suppresses intra-class noises between memory patterns. Overall, memory retrieval in the network progresses from rough to fine. Our model agrees with the push-pull phenomenon observed in neural data and sheds light on our understanding of the role of feedback in neural information processing.

Biography: Dr. Si Wu is Professor at School of Electronics Engineering & Computer Science, Principle Investigator at IDG/McGovern Institute for Brain Research, and Principle Investigator at PKU-Tsinghua Center for Life Science in Peking University. He was originally trained as a theoretical physicist and received his BSc, MSc, and PhD degrees all from Beijing Normal University (87-95). His research interests have turned to Artificial Intelligence and Computational Neuroscience since graduation. He worked as Postdocs at Hong Kong University of Science & Technology (95-97), Limburg University of Belgium (97-98), and Riken Brain Science Institute of Japan (98-00), and as Lecturer/Senior Lecturers at Sheffield University (00-02) and Sussex University (03-08) of UK. He came back to China in 2008, and worked as PI at Institute of Neuroscience in Chinese Academy of Sciences (08-11) and Professor in Beijing Normal University (11-18). His research interests focus on Computational Neuroscience and Brain-inspired Computing. He has published more than 100 papers, including top journals in neuroscience, such as Neuron, Nature Neuroscience, PNAS, J. Neurosci., and top conferences in AI, such as NIPS. He is now Co-editor-in-chief of Frontiers in Computational Neuroscience.      

 

Sen Song

Tsinghua University

Title: Recent progress in brain research and inspirations for neurocomputing

Abstract: Recently, big scale neuronal recordings are starting to reveal the way information is represented in the nervous system. At the same time, analysis of artificial neural networks trained by deep learning is also starting to reveal its representations. In this talk, I will try to summarize and compare representations in deep neural networks and the brain, regarding objects, object features, object relations, tree like structures and graph-like structures, and start to build a mathematical framework to describe them.

 Biography: Dr. Sen Song is an principal investigator at Tsinghua Laboratory for Brain and Intelligence and Department of Biomedical Engineering at Tsinghua University. He received his Ph.D. degree in Neuroscience from Brandeis University in 2002. Before joining Tsinghua in 2010, he did post-doctoral research at Cold Spring Harbor Laboratory and Massachusetts Institute of Technology. His work in computational neuroscience on spike-timing dependent plasticity and motif analysis of cortical connectivity have been widely cited and form some of the theoretical foundations of brain-inspired computing. His current work involves computational neuroscience, neural circuits underlying emotions and motivations, and the interface between neuroscience and artificial intelligence.

 

Wenming Zheng

Southeast University

Title: Action Intention Understanding and Emotion Recognition for Human Computer Interface

 Abstract: Action intention understanding and emotion recognition play an important role in human computer interface. In this talk, I will address the methods of action intention understanding and emotion recognition from psychophysiological signals, such as EEG or audiovisual signals. Then, I will also briefly address the applications of this research in medical treatment and education.

Biography: Wenming Zheng received his PhD degree in signal and information processing from the Department of Radio Engineering, Southeast University, Nanjing, China, in 2004. He is currently a Professor and the Director of the Key Laboratory of Child Development and Learning Science, Southeast University. He ever worked as a visiting scholar or visiting professor at Microsoft Research Asia (MSRA), Chinese University of Hong Kong (CUHK), University of Illinois at Urbana-Champaign (UIUC), and Cambridge University, respectively. His current research interests include affective information processing for multi-modal signals, e.g., facial expression, speech, and EEG signals, and their applications in education and medical care. Dr. Zheng was an Awardee the Microsoft Young Professor Professorship. He won the Second Prize of the National Technological Invention in 2018, the Second Prize of the Natural Science of Ministry of Education in 2008 and 2015, the Second Prize of the Jiangsu Provincial Science and Technology Progress in 2009. He served as an Associated Editor of several peer reviewed journals, such as IEEE TRANSACTIONS ON AFFECTIVE COMPUTING, Neurocomputing, and The Visual Computer. He is a Council Member of the Chinese Society of Cognitive Science.

Yijun Wang

Chinese Academy of Sciences

Title: Recent Progress in Brain-Machine Integration Technology

Abstract: The brain-computer interface (BCI) technology establishes a direct communication channel between the brain and external devices, which can replace, restore or enhance human’s perception, cognition and motor functions. In recent years, as a new form of hybrid intelligence, the BCI-based brain-machine integration technology has shown great potential in the fields of healthcare, human-computer interaction, and national defense. In this talk, I will introduce recent progress in the development of the brain-machine integration technology. I will first review the history, current status, methodology, and challenges in this field. I will then present examples of progress of the brain-machine integration technology in communication and control, human augmentation, multi-modal integration, and biometrics.

Biography: Yijun Wang is a Research Fellow at the Institute of Semiconductors, Chinese Academy of Sciences, and a member of CAS Center for Excellence in Brain Science and Intelligence Technology. He was selected by the Thousand Youth Talents Plan of China in 2015. He received a B.E. degree and a Ph.D. degree in biomedical engineering from Tsinghua University in 2001 and 2007, respectively. From 2008 to 2015, he was first a Postdoctoral Fellow and later an Assistant Project Scientist at the Institute for Neural Computation, University of California San Diego, USA. His research mainly focuses on neural engineering and neural computation. His research interests include brain-computer interface (BCI), biomedical signal processing, and machine learning. He has published more than 100 papers in scientific journals and conferences such as PNAS, Journal of Neuroscience, IEEE Transactions on Biomedical Engineering. His papers have been cited more than 4500 times according to Google Scholar.

Xiaolin Hu

Tsinghua University

Title: Deep Learning Predicts Correlation between a Functional Signature of Higher Visual Areas and Sparse Firing of Neurons

Abstract: Visual information in the visual cortex is processed in a hierarchical manner. Recent studies show that higher visual areas, such as V2, V3, and V4, respond more vigorously to images with naturalistic higher-order statistics than to images lacking them. This property is a functional signature of higher areas, as it is much weaker or even absent in the primary visual cortex (V1). However, the mechanism underlying this signature remains elusive. We studied this problem using computational models. In several typical hierarchical visual models including the AlexNet, VggNet and SHMAX, this signature was found to be prominent in higher layers but much weaker in lower layers. By changing both the model structure and experimental settings, we found that the signature strongly correlated with sparse firing of units in higher layers but not with any other factors, including model structure, training algorithm (supervised or unsupervised), receptive field size, and property of training stimuli. The results suggest an important role of sparse neuronal activity underlying this special feature of higher visual areas.

Biography: Xiaolin Hu is an associate professor in the Department of Computer Science and Technology, Tsinghua University, Beijing, China. He got his PhD degree in Automation and Computer-Aided Engineering at The Chinese University of Hong Kong in 2017. He was a postdoc at Tsinghua University during 2017-2019. His research areas include artificial neural networks and computational neuroscience. His main research interests include developing brain-inspired computational models and revealing the visual and auditory information processing mechanism in the brain. He has published over 70 research papers in journals include IEEE Transactions on Neural Networks and Learning Systems, IEEE Transactions on Image Processing, IEEE Transactions on Cybernetics, PLoS Computational Biology, Neural Computation, and conferences include CVPR, NIPS, AAAI. He serve as an Associate Editor of IEEE Transactions on Neural Networks and Learning Systems and Associate Editor of Cognitive Neurodynamics.

Jian Liu

University of Leicester

Title: Towards the next generation of computer vision: visual computation with spikes

Abstract: Neuromorphic computing has been suggested as the next generation of computational strategy.  In terms of vision, the retina is the first stage of visual processing in the brain. The retinal coding is for understanding how the brain processes stimulus from the environment, moreover, it is also a cornerstone for designing algorithms of visual coding, where encoding and decoding of incoming stimulus are needed for better performance of physical devices. Here, by using the retina as a model system, we develop some data-driven approaches, spike-triggered non-negative matrix factorization and deep learning nets for characterizing the encoding and decoding of natural scenes by retinal neuronal spikes. I further demonstrate how these computational principles of neuroscience can be transferred to neuromorphic chips for the next generation of the artificial retina. As a proof of concept, the revealed mechanisms and proposed algorithms here for the retinal visual processing can provide new insights into neuromorphic computing with the signal of events or neural spikes.

Biography: Dr. Jian Liu received the Ph.D. in mathematics from UCLA, then worked as Postdoc Fellow at CNRS, France, and University of Goettingen, Germany. He is currently a Lecturer of Computational Neuroscience at University of Leicester, UK. His area of research includes computational neuroscience and brain-inspired computation for artificial intelligence. His work was published in Nature communications, eLife, Journal of neuroscience, PLoS computational biology, IEEE Transactions on Neural Networks and Learning Systems, IEEE Transactions on cybernetics, etc.

雷鋒網(wǎng)報(bào)道。

雷峰網(wǎng)原創(chuàng)文章，未經(jīng)授權(quán)禁止轉(zhuǎn)載。詳情見(jiàn)轉(zhuǎn)載須知。