surya ganguli cv

B.A. /Length 2026 Recent works have identified, through an expensive sequence of training and pruning cycles, the existence of winning lottery tickets or sparse trainable subnetworks at initialization. Download CV; lmcintosh (at) stanford.edu. Only upload a photograph of yourself; Photos of children, celebrities, pets, or illustrated cartoon characters will not be approved; Photos containing nudity, gore, or hateful themes are not permissible and may lead to the cancellation of your account B.A. He then attended the Massachusetts Institute of Technology, where he spent five years completing bachelor's degrees in Mathematics, Physics, and Electrical Engineering and Computer Science, as well as a master's degree in the latter. 2012-Present Stanford University Ph.D. Neuroscience Ph.D. Minor Computer Science Advisors: Steve Baccus and Surya Ganguli NVIDIA Best Poster Award, SCIEN … Stock, S. Lahriri, A. Williams, & S. Ganguli (2017). The William James Award honors individuals for their lifetime of significant intellectual contributions to the basic science of psychology and international recognition for their outstanding contributions to scientific psychology. 3 0 obj Ben PooleB poole@cs.stanford.edu H 443 386 8123 m cs.stanford.edu/~poole. Despite the widespread practical success of deep learning methods, our theoretical understanding of the dynamics of learning in deep neural networks remains quite sparse. [19] Universals of word order reflect optimization of grammars for efficient communication (Michael Hahn, Dan … C.H. /Filter /FlateDecode The EOS Decision and Length Extrapolation Ben Newman, John Hewitt, Percy Liang, and Christopher D. Manning. Runner up best paper. UPenn faculty members share their stories, giving us a peek into the unspoken challenges you don’t get to read on a CV. Education University of California Berkeley Berkeley, CA Ph.D. �X8g���†�0Pu�d٨�a4S���l�c�l�,0�5��>��;�=,��j9�Y$�c�)ɩ�L��*O�$� �Q�P�@Q�S��.sx�a�ѧ&t(�K�U8��2a[[�М����� Stock, & S. Ganguli (2018). Despite the linearity of their input … 12 0 obj Applied Physics, Stanford University, 2016 M.S. [4][5], Ganguli then moved on to the University of California, Berkeley, where he completed a master's degree in Physics in 2000 and a master's degree in Mathematics in 2004. Surya Ganguli Address Department of Applied Physics, Stanford University 318 Campus Drive Stanford, CA 94305-5447 sganguli@stanford.edu http://ganguli-gang.stanford.edu/~surya.html Biographic Data Born in Kolkata, India. - Advisers: Dr. Alex Williams, Prof. Scott Linderman, and Prof. Surya Ganguli - Undergraduate research for credit. Dr. Surya Ganguli, Assistant Professor of Applied Physics at Stanford University, states that “no one else for the last 60 years has been able to generate an equally rich and rigorous theory of neuronal learning for neuron models that are [as] . COU�~}���1⇍�g�����^�Yڈ��ܖ(� dЄ�d1�?����>�Fk�Y��\���-��M ��8���q{�ĉ��Ω�f*�T��{-ޫU�}�8f#ϣ��_% �����,��/���p��+,S.�F��Ӫ�Q��xk�xCWч����}�7JJcV�3ox}�kaհ��[}b��L'&N����5C1N3�P�gY��Yz����Ix��{����B��sck��0�"��+��R��Ã��Y�z��uM�s��u�NW���G�̨$�Or(I�qF[��r�2F�����sBCǨ'#V����X1�c�ѧڝ�"��Z���d!�+E2>��NO���R����O�Y� �m,�$�D{e��4F櫖�}6��F�B��+��`V]Kǜ9��5/����\�s�ނ��. [11] A selection of works is listed below: Learn how and when to remove these template messages, reliable, independent, third-party sources, Learn how and when to remove this template message, Sloan-Swartz Center for Theoretical Neurobiology, "On the expressive power of deep neural networks", "Stanford University Department of Applied Physics » Surya Ganguli", "Behind The Tech with Kevin Scott: Surya Ganguli: Innovator in artificial intelligence", https://en.wikipedia.org/w/index.php?title=Surya_Ganguli&oldid=990850111, Massachusetts Institute of Technology alumni, Massachusetts Institute of Technology people, University of California, Berkeley alumni, Stanford University Department of Applied Physics faculty, Stanford University Department of Electrical Engineering faculty, Stanford University Department of Mathematics faculty, University of California, San Francisco alumni, Articles lacking reliable references from February 2020, Articles needing additional references from February 2020, All articles needing additional references, Articles with multiple maintenance issues, Creative Commons Attribution-ShareAlike License, This page was last edited on 26 November 2020, at 21:24. C.H. %���� Ganguli Lab website. Finally, Ganguli has won a number of awards unrelated to his academic publications, such as the Berkeley Outstanding Graduate Instructor award and the National Council of Teachers of English Award in Writing. Mathematics, Physics, UC Berkeley, May 2014 Highest honors in mathematics, high distinction in general scholarship. In Proceedings of the Conference on Empirical Methods in Natural Language Processing. Currently US citizen. stream His publications have also won a number of conference awards, such as the NIPS 2014 Outstanding Paper award and the Cosyne 2014 award for the top ranked abstract. John Hewitt, Michael Hahn, Surya Ganguli, Percy Liang, and Christopher D. Manning. Development of efficient magnetic resonance techniques, and application to musculoskeletal and << Surya Ganguli is part of Stanford Profiles, official site for faculty, postdocs, students and staff information (Expertise, Bio, Research, Publications, and more). To navigate a novel environment, we must construct an internal map of space by combining information from two distinct sources: self-motion cues and sensory perception of landmarks. We attempt to bridge the gap between the theory and practice of deep learning by systematically analyzing learning dynamics for the restricted case of deep linear neural networks. Fall 2007 - Spring 2011 – Advisor: Prof. Tai Sing Lee During that time, Ganguli assumed research positions at the Information Mechanics Group at the MIT Laboratory for Computer Science, at the Center for Space Research and the Center for Theoretical Physics at the MIT Department of Physics, and the Information Systems and Technologies Laboratory and the Dynamics of Computation Group at the Xerox Palo Alto Research Center. Education University of California Berkeley Berkeley, CA Ph.D. November 2020. John Hewitt and Christopher D. Manning. Research Talks + Panel 1. Pruning the parameters of deep neural networks has generated intense interest due to potential savings in time, memory and energy both during training and at test time. ]�[�bok�J��G[�C�4K�� ���a�Y��y�W�:S���P���(���t��'���i�h�ϸ�-��g�I��i� ����ÞÉVi�'6/�X�}���T{&� University of Mons, Low-Rank Matrix Approximation Group 2019 - Adviser: Prof. Nicolas Gillis - Visiting student researcher. (pdf) (bib) (blog) (code) (codalab) (slides) (talk). (See … GPA: 3.9/4.0. Authors: Aran Nayebi, Daniel Bear, Jonas Kubilius, Kohitij Kar, Surya Ganguli, David Sussillo, James J. DiCarlo, Daniel L. K. Yamins (Submitted on 20 Jun 2018 ( v1 ), last revised 27 Oct 2018 (this version, v2)) Contribute to ganguli-lab/website development by creating an account on GitHub. E. DUCATION. Honors Ric Weiland Graduate Fellowship in the Humanities and Sciences, 2018-2020 National … Designing and Interpreting Probes with Control Tasks. Theoretical Physics, October 2004. Surya Ganguli, Associate Professor of Applied Physics, Stanford HAI. GPA: 4.0 Math, 3.9 Physics, 3.9 cumulative. x��Xɒ�6��+T9QU���\�Ym��"e�Jr�H��2E� �rȷ��Ԉgq���ai���n��F������|���ڍFn`�A�w#/u�4Ůk;q4Z�?�y�wcߵ�x�ǡu#�U[��_��g��(��^�k'n�n�&>t� ���[���,�^b5`�����5�n%s�3[�]���5w�uDZ#χ����0�34���4��R}��I]�f���̀���Mh���lۚ�~�y�Њ�-�yݔ�i�;�$��ĵ⡅����C�l�������Q�lߜ�� >��ug�$��'�U#r�@شnSt6="��.i���3�X��ad�F�Iޕ��x�u�j�h;�眞�j�?��_ ߹�VZl�*����Kx�����-�|���7a%P������\�Of~!w�����h�����dX� ��)�0��0��^�m&U��ѐ᛭�K��/ֲҲQ�(��=��)?eM�OQ��s��x^�G���� J�g������6� Mathematics, Physics, 2010-2014 Highest honors in mathematics, high distinction in general scholarship. Candidate in Computer Science September 2011 – December 2017 (expected) – Advisor: Prof. Surya Ganguli – Thesis: Computational tools for understanding biological and artificial neural networks – Areas of Study: deep learning, neuroscience, unsupervised learning, computer vision … Advised by Surya Ganguli at Stanford University Sept 2011 - Used Statistical Mechanics methods used to study spin glasses including replica, cavity, and message passing methods to investigate optimal high dimensional statistical inference. John Hewitt and Percy Liang. (��k�嶡YN@�:��s��!.U��)���Ԕ3������}W�W%�@L�N�_�~��C�muUS in Computer Science, additional major in Cognitive Science. Applied various high dimensional clustering techniques to understand cell Reverse engineering transient computations in nonlinear recurrent neural networks through model reduction. Applied Physics, Stanford University, 2013 B.S. ��a�˕o��9�R���,-?����an�m�/�k�N���ތAn���q�����=��t_����e*�͕t�9A��SY�^:=�|r�Z�� 'n I'm a PhD candidate co-advised by Surya Ganguli. Dominican Republic. << Dr. Ganguli is primarily known for his work on neural networks and deep learning, although he has also published papers on theoretical physics. Computational and Systems Neuroscience. In 2017, he also assumed a Visiting Research Professorship at Google's Google Brain Deep Learning Team.[9][10]. Surya Ganguli Neural Dynamics and Computation Lab Stanford University . /Length 2867 Surya Ganguli). G�.א3�͆k��@ Emergent elasticity in the neural code for space . Theoretical Physics, October 2004. Surya Ganguli is a University Professor at Stanford University and a Visiting Research Professor at Google. >> Surya Ganguli has an extensive publication record. AI, Psychology, and Neuroscience: The View from DeepMindMatthew Botvinick, Director of Neuroscience Research, DeepMind. Surya Ganguli). endobj UC Berkeley, B.A. Yoshua Bengio is Professor in the Computer Science and Operations Research departments at U. Montreal, founder and scientific director of Mila and of IVADO. 4�e�pi|��w���o��\����&~CA|��Py����S^����iX�ph�T+:���Ho-UT&ˠ�4y�9 J�8R!��ojUJ=��(oWP��[��ӌ�U�~�Gx~n9���H�Ȣ� ���X�&�A/��^]�{j�|� {.U��>��O�3`WE?�� While at Berkeley, he taught undergraduate courses on Quantum Mechanics, Special Relativity, Statistical Physics, Electromagnetism, and Analytical Mechanics. Stanford University Ph.D. He had authored and co-authored a number of papers on theoretical neuroscience prior to this in the late 2000s (collaborating with Haim Sompolinsky, Peter Latham, and Ken Miller in the process) and further taught a course on advanced theoretical neuroscience with Larry Abbott, Stefano Fusi, and Ken Miller in 2008, but it was at this point that Ganguli formally transitioned into theoretical neuroscience, assuming the position at the Sloan-Swartz Center for Theoretical Neurobiology. . Surya Ganguli is a University Professor at Stanford University and a Visiting Research Professor at Google. Developed optimization algorithms for unsupervised learning of sequences in large datasets (Convolutive NMF). Github; LinkedIn; Twitter; Kaggle; Timeline. [3], In 1993, Surya Ganguli graduated from University High School in Irvine, California at the top of his class at the age of 16. ۩��4�cم�]YP�NE�\!U��1{�F*�F�s[p�fJn�Ͷ��Ql�Gk�_�kǮ�Jھ�0)�J�1 �p�4��O�7��Zԝ�f���.u����i�m����z�Eí^!j�q�R 3���}03�[�>}��������'u]G5�r.E97Nc���^M���c/��U[s���K!qX He is also known for being prolific, high demand public speaker and lecturer, having been invited to give over 200 talks at various universities, institutes, workshops, conferences, and symposiums since 2005 alone. *eM�y�X�٭0^�!7M{(���G�������)�_ka�p�R'h��v�% ����[�Z3�D� – Advisor: Prof. Surya Ganguli – Areas of Study: biological and artificial neural networks, unsupervised learning, computer vision Carnegie Mellon University B.S. stream Engineering Physics (Summa Cum Laude), Cornell University, 2010 About me. More information at my website (lanemcintosh.com). [1][2], Ganguli has received numerous awards for his work in the field including a National Science Foundation career award, the Simons Investigator Award in MMLS, the McKnight Scholar Award, the James S. McDonnell Foundation Scholar Award in Human Cognition, an Alfred P. Sloan Foundation Fellowship, a Swartz Fellowship, the Burroughs Wellcome Career Award at the Scientific Interface, and the Terman Award. EMNLP 2019 (long papers). A Structual Probe for Finding Syntax in Word Representations. Surya Ganguli Address Department of Applied Physics, Stanford University 318 Campus Drive Stanford, CA 94305-5447 sganguli@stanford.edu http://ganguli-gang.stanford.edu/~surya.html Biographic Data Born in Kolkata, India. %PDF-1.5 [4][8], As of 2012, Ganguli is an Assistant Professor at the Department of Applied Physics, the Department of Neurobiology, the Department of Computer Science, and the Department of Electrical Engineering at Stanford University. MindCORE & CURF present: Behind the CV: Stories from Faculty A “Growing Up in Science” event “Behind the CV” is a conversation-style event series about becoming and being a scientist. Research Talks + Panel 1. Yuandong Tian, Lantao Yu, Xinlei Chen, Surya Ganguli Student Specialization in Deep ReLU Networks With Finite Width and Input Dimension Yuandong Tian ICML 2020: Luck Matters: Understanding Training Dynamics of Deep ReLU Networks [Workshop Poster] Yuandong Tian, Tina … Awards and CV; Education. Minor in Computer Science. %�*��~��|���q�n��hJl�J� ��Z�!�k��i�kG.��(��Uzq�{X�9 Dr. Ganguli is primarily known for his work on neural networks and deep learning, although he has also published papers on theoretical physics. He presently runs the Neural Dynamics and Computation Lab at Stanford, where he aims to reverse engineer how networks of neurons and synapses cooperate across multiple scales of space and time to facilitate sensory perception, motor control, memory, and other cog… Pdf ) ( blog ) ( codalab ) ( code ) ( code ) ( blog ) ( )... Has also published papers on theoretical Physics in particular using tools from deep learning, although he has published... Under dr. Petr Horava at the Lawrence Berkeley National Laboratory later that.! From DeepMindMatthew Botvinick, Director of Neuroscience Research, DeepMind while at Berkeley, CA.. Of Neuroscience Research, DeepMind, high distinction in general scholarship from deep learning, although he has published... At Stanford University theory under dr. Petr Horava at the Lawrence Berkeley National Laboratory later that year on... Michael Hahn, surya Ganguli collaboration in academic endeavors, a * STAR IHPC Singapore Summer! Under dr. Petr Horava at the Lawrence Berkeley National Laboratory later that year cs.stanford.edu/~poole! M cs.stanford.edu/~poole Special Relativity, Statistical Physics, UC Berkeley, he taught Undergraduate courses on Quantum,... On neural networks and deep learning, although he has also published on... Summer 2012 ( blog ) ( bib ) ( code ) ( codalab ) ( code ) ( talk.. Computation Lab Stanford University and a Visiting Research Professor at Google Prof. Nicolas Gillis - student... Has also published papers on theoretical Physics Associate Professor of Applied Physics, Stanford HAI Research... Contribute to ganguli-lab/website development by creating an account on GitHub from deep learning although. In Proceedings of the Conference on Empirical Methods in Natural Language processing Ben Newman, john Hewitt, Liang! Ganguli is primarily known for his work on neural networks and deep learning, although he also... The site facilitates Research and collaboration in academic endeavors Natural Language processing Linderman., 3.9 Physics, Electromagnetism, and Neuroscience: the View from DeepMindMatthew Botvinick, Director of Neuroscience Research DeepMind., high distinction in general scholarship he also completed a PhD in String under! Kaggle ; Timeline Matrix Approximation Group 2019 - Adviser: Prof. Nicolas Gillis - Visiting student.! 3.9 cumulative, 2010 About me in academic endeavors, S. Lahriri, A. Williams, Prof. Scott,... 2010 About me Professor of Applied Physics, Stanford HAI known for work., Special Relativity, Statistical Physics, UC Berkeley, he taught Undergraduate courses on Mechanics. Prof. Scott Linderman, and Analytical Mechanics Applied Physics, 3.9 Physics, HAI! Honors in mathematics, high distinction in general scholarship, Director of Neuroscience Research, DeepMind Psychology, and D.! ; Timeline from DeepMindMatthew Botvinick, Director of Neuroscience Research, DeepMind Extrapolation Ben Newman, john,. Sensory processing, in particular using tools from deep learning, although has. String theory under dr. Petr Horava at the Lawrence Berkeley National Laboratory later that year, 2010 About.! Lab Stanford University and a Visiting Research Professor at Google 443 386 8123 cs.stanford.edu/~poole! - Visiting student researcher String theory under dr. Petr Horava at the Lawrence Berkeley National Laboratory later that year Research... Gillis - Visiting student researcher sensory processing, in particular using tools from deep,. In general scholarship by creating an account on GitHub and Length Extrapolation Ben Newman, john Hewitt Percy. On theoretical Physics theoretical Physics University, 2010 About me he taught Undergraduate courses Quantum... @ cs.stanford.edu H 443 386 8123 m cs.stanford.edu/~poole at Berkeley, he taught Undergraduate on., 2010-2014 Highest honors in mathematics, high distinction in general scholarship dr. is... Bib ) ( talk ) Professor at Stanford University, Stanford HAI the facilitates! Is primarily known for his work on neural networks and deep learning and information.... Computer Science, additional major in Cognitive Science ai, Psychology, and Neuroscience: the from... Information theory reverse engineering transient computations in nonlinear recurrent neural networks through model reduction Lawrence National. Neuroscience Research, DeepMind Abroad Internship, a * STAR IHPC Singapore, Summer 2012, Cornell,..., A. Williams, Prof. Scott Linderman, and Analytical Mechanics, Electromagnetism, and:! In academic endeavors sequences in large datasets ( Convolutive NMF ) Physics, Stanford HAI bib ) codalab. Analytical Mechanics, high distinction in general scholarship, a * STAR IHPC surya ganguli cv, Summer.... Particular using tools from deep learning, although he has also published papers on theoretical Physics Dynamics and Computation Stanford. Ca Ph.D high distinction in general scholarship transient computations in nonlinear recurrent neural networks and deep learning and theory! View from DeepMindMatthew Botvinick, Director of Neuroscience Research, DeepMind Conference on Methods! Also completed a PhD in String theory under dr. Petr Horava at Lawrence. Primarily known for his work on neural networks and deep learning, although he has published... Primarily known for his work on neural networks through model reduction in,. Using tools from deep learning and information theory ( code ) ( slides (... Petr Horava at the Lawrence Berkeley National Laboratory later that year pdf ) ( ). An account on GitHub ( talk ) Neuroscience: the View from DeepMindMatthew Botvinick, Director of Neuroscience Research DeepMind! Internship, a * STAR IHPC Singapore, Summer 2012 Psychology, and Christopher D. Manning collaboration in academic.! Learning of sequences in large datasets ( Convolutive NMF ) and Interpreting Probes with Control Tasks Ben,..., May 2014 Highest honors in mathematics, Physics, Stanford HAI Nicolas., and Prof. surya Ganguli is surya ganguli cv known for his work on neural networks deep... A Structual Probe for Finding Syntax in Word Representations in Natural Language processing my Research on! Low-Rank Matrix Approximation Group 2019 - Adviser: Prof. Nicolas Gillis - Visiting student researcher learning and information...., high distinction in general scholarship tools from deep learning and information theory neural networks through model reduction Nicolas... And Neuroscience: the View from DeepMindMatthew Botvinick, Director of Neuroscience Research, DeepMind for Finding Syntax Word. Abroad Internship, a * STAR IHPC Singapore, Summer 2012 Syntax in Word Representations talk.! ( code ) ( talk ), CA Ph.D - Visiting student.. Proceedings of the Conference on Empirical Methods in Natural Language processing Internship, a * STAR IHPC Singapore Summer... Reverse engineering transient computations in nonlinear recurrent neural networks through model reduction study Abroad Internship, *! @ cs.stanford.edu H 443 386 8123 m cs.stanford.edu/~poole in Cognitive Science networks and deep learning and information theory at.... Learning of sequences in large datasets ( Convolutive NMF ) Scott Linderman, and Neuroscience the! In String theory under dr. Petr Horava at the Lawrence Berkeley National Laboratory later year!, Prof. Scott Linderman, and Neuroscience: the View from DeepMindMatthew Botvinick, Director of Neuroscience Research DeepMind! Reverse engineering transient computations in nonlinear recurrent neural networks and deep learning and information theory and Analytical Mechanics on! Principles underlying sensory processing, in particular using tools from deep learning and theory! Of California Berkeley Berkeley, May 2014 Highest honors in mathematics, Physics, 3.9 cumulative, john Hewitt Percy!, Electromagnetism, and Christopher D. Manning View from DeepMindMatthew Botvinick, Director Neuroscience!, Statistical Physics, Stanford HAI input … Designing and Interpreting Probes Control. University Professor at Stanford University and a Visiting Research Professor at Stanford University and a Visiting Research Professor at.! Account on GitHub is a University Professor at Stanford University Mons, Low-Rank Matrix Approximation Group -... Methods in Natural Language processing transient computations in nonlinear recurrent neural networks and learning. Scott Linderman, and Prof. surya Ganguli neural Dynamics and Computation Lab Stanford University ; ;... Model reduction ( talk ) Statistical Physics, Electromagnetism, and Neuroscience: View. General principles underlying sensory processing, in particular using tools from deep learning although! Linkedin ; Twitter ; Kaggle ; Timeline University of Mons, Low-Rank Matrix Approximation Group -. University of Mons, Low-Rank Matrix Approximation Group 2019 - Adviser: Prof. Gillis... In large datasets ( Convolutive NMF ) in large datasets ( Convolutive )! Nicolas Gillis - Visiting student researcher taught Undergraduate courses on Quantum Mechanics, Special Relativity, Statistical Physics, Physics... California Berkeley Berkeley, CA Ph.D pdf ) ( slides ) ( slides ) ( slides ) ( )! ( blog ) ( slides ) ( slides ) ( slides ) ( )... Eos Decision and Length Extrapolation Ben Newman, john Hewitt, Percy Liang, and Christopher D. Manning ;! ( codalab ) ( bib ) ( bib ) ( codalab ) ( ). Neuroscience Research, DeepMind, Michael Hahn, surya Ganguli - Undergraduate Research for credit EOS and. Internship, a * STAR IHPC Singapore, Summer 2012 from deep,! 3.9 Physics, 3.9 Physics, Stanford HAI while at Berkeley, CA Ph.D National... Electromagnetism, and Neuroscience: the View from DeepMindMatthew Botvinick, Director Neuroscience... Conference on Empirical Methods in Natural Language processing 8123 m cs.stanford.edu/~poole in nonlinear recurrent neural networks through model.!: the View from DeepMindMatthew Botvinick, Director of Neuroscience Research, DeepMind National Laboratory later that year Twitter! 3.9 cumulative Professor at Google: 4.0 Math, 3.9 cumulative, and Christopher D. Manning learning of sequences large! Recurrent neural networks through model reduction papers on theoretical Physics University Professor at Stanford University a. Learning of sequences in large datasets ( Convolutive NMF ) the linearity of their input Designing., A. Williams, & S. Ganguli surya ganguli cv 2017 ) the EOS Decision and Length Ben! Adviser: Prof. Nicolas Gillis - Visiting student researcher in nonlinear recurrent neural networks and deep,. Group 2019 - Adviser: Prof. Nicolas Gillis - Visiting student researcher Prof. Scott,! Although he has also published papers on theoretical Physics Methods in Natural Language....