View Production Programme [PDF]
Total Page:16
File Type:pdf, Size:1020Kb
Load more
Recommended publications
-
Complex Internal Representations in Sensorimotor Decision Making: a Bayesian Investigation Doctor of Philosophy, 2014
COMPLEXINTERNALREPRESENTATIONS INSENSORIMOTORDECISIONMAKING: A BAYESIAN INVESTIGATION luigi acerbi I V N E R U S E I T H Y T O H F G E R D I N B U Doctor of Philosophy Doctoral Training Centre for Computational Neuroscience Institute of Perception, Action and Behaviour School of Informatics University of Edinburgh 2014 Luigi Acerbi: Complex internal representations in sensorimotor decision making: A Bayesian investigation Doctor of Philosophy, 2014 supervisors: Prof. Sethu Vijayakumar, Ph.D., FRSE Prof. Daniel M. Wolpert, D.Phil., FRS, FMedSci ABSTRACT The past twenty years have seen a successful formalization of the idea that percep- tion is a form of probabilistic inference. Bayesian Decision Theory (BDT) provides a neat mathematical framework for describing how an ideal observer and actor should interpret incoming sensory stimuli and act in the face of uncertainty. The predictions of BDT, however, crucially depend on the observer’s internal models, represented in the Bayesian framework by priors, likelihoods, and the loss function. Arguably, only in the simplest scenarios (e.g., with a few Gaussian variables) we can expect a real observer’s internal representations to perfectly match the true statistics of the task at hand, and to conform to exact Bayesian computations, but how humans systemati- cally deviate from BDT in more complex cases is yet to be understood. In this thesis we theoretically and experimentally investigate how people represent and perform probabilistic inference with complex (beyond Gaussian) one-dimensional distributions of stimuli in the context of sensorimotor decision making. The goal is to reconstruct the observers’ internal representations and details of their decision- making process from the behavioural data – by employing Bayesian inference to un- cover properties of a system, the ideal observer, that is believed to perform Bayesian inference itself. -
Moira Rose (Molly) Dillon
Moira Rose Dillon Department of Psychology, New York University, 6 Washington Place, New York, NY 10003 Email – [email protected] Departmental Website – http://as.nyu.edu/psychology/people/faculty.Moira-Dillon.html Lab Website – https://www.labdevelopingmind.com Employment New York University New York, NY Assistant Professor, Department of Psychology, Faculty of Arts and Sciences (July 2017-present) Concurrent Positions New York University New York, NY Faculty Affiliate, Institute of Human Development and Social Change, Steinhardt School of Culture, Education, and Human Development (May 2019-present) Massachusetts Institute of Technology Cambridge, MA Invited Researcher, Abdul Latif Jameel Poverty Action Lab (J-PAL), Foundations of Learning (April 2021-present) Education Harvard University Cambridge, MA (August 2011-May 2017) Ph.D., Psychology (May 2017) A.M., Psychology (May 2014) Yale University New Haven, CT (August 2004-May 2008) B.A., Cognitive Science; Art (May 2008) Funding 2019-2024 National Science Foundation (PI: $1,718,437) CAREER: Becoming Euclid: Characterizing the geometric intuitions that support formal learning in mathematics (PI: $24,671) CLB: Career-Life Balance Faculty Early Career Development Program Supplement 2019-2023 DARPA (Co-PI; to NYU: $1,703,553; to Dillon: $871,874) Cognitive milestones for Machine Common Sense, Co-PI: Brenden Lake 2019-2021 Jacobs Foundation (PI: 150,000 CHF) Early Career Research Fellowship 2018-2019 Institute of Human Development and Social Change at NYU (PI: $14,848) The arc of geometric -
About Krai About Olga Bell
Tuesday, June 23 at 8:30pm at Pier 15 ABOUT KRAI Co-presented by Lower Manhattan Cultural Council and New Amsterdam Presents as part of the River To River Festival 2015. Krai (край) is the Russian word for edge, limit, frontier or hinterland. Present-day Russia is divided into a myriad of ‘federal subjects’, including nine krais. In this capacity the term is a political designation, like ‘territory’, but for the earliest Russians these places represented both the promise and terror of the vast unknown. While much has been written about Russia’s major cities, Olga Bell’s Krai is a composition dedicated to the rest of the map: the wilderness, the towns, the inhabitants and their stories. From the Cossack melodies of Krasnodar Krai in the West to the Chukchi drumming of Kamchatka Krai in the Russian Far East, Krai is a forty-minute journey across the Eurasian landmass. The live incarnation of Krai features Olga plus five vocalists singing entirely in Russian, supported by a electro-acoustic chamber ensemble of cello, electric guitar, bass, pitched drums, mallet percussion and electronics. The setting for this performance of Krai is Pier 15, where a video installation showing real and imagined landscapes from each territory completes a unique and immersive experience of Olga Bell’s groundbreaking, cross-cultural work. ABOUT OLGA BELL Born in Moscow and raised in Alaska, Olga Bell is a New York-based composer, producer and performer of intrepid, aurally omnivorous music. While much has been written about Russia’s major cities, she’s concerned with the rest of the map: the wilderness, the towns, the inhabitants and their stories. -
Why the Bayesian Brain May Need No Noise
Stochasticity from function – why the Bayesian brain may need no noise Dominik Dolda,b,1,2, Ilja Bytschoka,1, Akos F. Kungla,b, Andreas Baumbacha, Oliver Breitwiesera, Walter Sennb, Johannes Schemmela, Karlheinz Meiera, Mihai A. Petrovicia,b,1,2 aKirchhoff-Institute for Physics, Heidelberg University. bDepartment of Physiology, University of Bern. 1Authors with equal contributions. 2Corresponding authors: [email protected], [email protected]. August 27, 2019 Abstract An increasing body of evidence suggests that the trial-to-trial variability of spiking activity in the brain is not mere noise, but rather the reflection of a sampling-based encoding scheme for probabilistic computing. Since the precise statistical properties of neural activity are important in this context, many models assume an ad-hoc source of well-behaved, explicit noise, either on the input or on the output side of single neuron dynamics, most often assuming an independent Poisson process in either case. However, these assumptions are somewhat problematic: neighboring neurons tend to share receptive fields, rendering both their input and their output correlated; at the same time, neurons are known to behave largely deterministically, as a function of their membrane potential and conductance. We suggest that spiking neural networks may have no need for noise to perform sampling-based Bayesian inference. We study analytically the effect of auto- and cross-correlations in functional Bayesian spiking networks and demonstrate how their effect translates to synaptic interaction strengths, rendering them controllable through synaptic plasticity. This allows even small ensembles of interconnected deterministic spiking networks to simultaneously and co-dependently shape their output activity through learning, enabling them to perform complex Bayesian computation without any need for noise, which we demonstrate in silico, both in classical simulation and in neuromorphic emulation. -
Wilken and Ma 2004
Journal of Vision (2004) 4, 1120-1135 http://journalofvision.org/4/12/11/ 1120 A detection theory account of change detection Division of Biology, California Institute of Technology, Patrick Wilken Pasadena, CA, USA Division of Biology, California Institute of Technology, Wei Ji Ma Pasadena, CA, USA Previous studies have suggested that visual short-term memory (VSTM) has a storage limit of approximately four items. However, the type of high-threshold (HT) model used to derive this estimate is based on a number of assumptions that have been criticized in other experimental paradigms (e.g., visual search). Here we report findings from nine experiments in which VSTM for color, spatial frequency, and orientation was modeled using a signal detection theory (SDT) approach. In Experiments 1-6, two arrays composed of multiple stimulus elements were presented for 100 ms with a 1500 ms ISI. Observers were asked to report in a yes/no fashion whether there was any difference between the first and second arrays, and to rate their confidence in their response on a 1-4 scale. In Experiments 1-3, only one stimulus element difference could occur (T = 1) while set size was varied. In Experiments 4-6, set size was fixed while the number of stimuli that might change was varied (T = 1, 2, 3, and 4). Three general models were tested against the receiver operating characteristics generated by the six experiments. In addition to the HT model, two SDT models were tried: one assuming summation of signals prior to a decision, the other using a max rule. In Experiments 7-9, observers were asked to directly report the relevant feature attribute of a stimulus presented 1500 ms previously, from an array of varying set size. -
Using Games to Understand Intelligence
Using Games to Understand Intelligence Franziska Brandle¨ 1, Kelsey Allen2, Joshua B. Tenenbaum2 & Eric Schulz1 1Max Planck Research Group Computational Principles of Intelligence, Max Planck Institute for Biological Cybernetics 2Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology Over the last decades, games have become one of the most isons between human and machine agents, for example in re- popular recreational activities, not only among children but search of human tool use (Allen et al., 2020). Games also also among adults. Consequently, they have also gained pop- offer complex environments for human-agent collaborations ularity as an avenue for studying cognition. Games offer sev- (Fan, Dinculescu, & Ha, 2019). Games have gained increas- eral advantages, such as the possibility to gather big data sets, ing attention during recent meetings of the Cognitive Science engage participants to play for a long time, and better resem- Society, with presentations on multi-agent collaboration (Wu blance of real world complexities. In this workshop, we will et al., 2020), foraging (Garg & Kello, 2020) and mastery (An- bring together leading researchers from across the cognitive derson, 2020) (all examples taken from the 2020 Conference). sciences to explore how games can be used to study diverse aspects of intelligent behavior, explore their differences com- Goal and scope pared to classical lab experiments, and discuss the future of The aim of this workshop is to bring together scientists who game-based cognitive science research. share a joint interest in using games as a tool to research in- Keywords: Games; Cognition; Big Data; Computation telligence. We have invited leading academics from cogni- tive science who apply games in their research. -
Wei Ji Ma CV Apr 2021 Center for Neural Science
Wei Ji Ma CV Apr 2021 Center for Neural Science and Department of Psychology New York University 4 Washington Place, New York, NY 10003, USA, (212) 992 6530 [email protected] Lab website: http://www.cns.nyu.edu/malab POSITIONS 2020-present Professor, Center for Neural Science and Department of Psychology, New York University ● Affiliate faculty in the Institute for Decision-Making, the Center for Data Science, the Neuroscience Institute, and the Center for Experimental Social Science ● Collaborating Faculty of the NYU-ECNU Institute of Brain and Cognitive Science at NYU Shanghai 2013-2020 Associate Professor, Center for Neural Science and Department of Psychology, New York University (with affiliations as above) 2008-2013 Assistant Professor, Department of Neuroscience, Baylor College of Medicine ● Adjunct faculty in the Department of Psychology, Rice University TRAINING 2004-2008 Postdoc, Department of Brain and Cognitive Sciences, University of Rochester. Advisor: Alexandre Pouget 2002-2004 Postdoc, Division of Biology, California Institute of Technology. Advisor: Christof Koch 1996-2001 PhD in Theoretical Physics, University of Groningen, the Netherlands Jan-Jun 2000 Visiting PhD student, Department of Physics, Princeton University. Advisor: Erik Verlinde 1994-1997 BS/MS in Mathematics, University of Groningen 1993-1996 BS/MS in Physics, University of Groningen RESEARCH INTERESTS Planning, reinforcement learning, social cognition, perceptual decision-making, perceptual organization, visual working memory, comparative cognition, optimality/rationality, approximate inference, probabilistic computation, efficient coding, computational methods. TEACHING, MENTORING, AND OUTREACH Improving the quality of undergraduate teaching, improving mentorship, diversity/equity/inclusion, science outreach, science writing, science advocacy, translating science to policy, social justice. Founder of Growing up in science Founding member of the Scientist Action and Advocacy Network (ScAAN) Founding member of NeuWrite NYU 1 PUBLICATIONS 1. -
COSYNE 2012 Workshops February 27 & 28, 2012 Snowbird, Utah
COSYNE 2012 Workshops February 27 & 28, 2012 Snowbird, Utah Monday, February 27 Organizer(s) Location 1. Coding and Computation in visual short-term memory. W.J. Ma Superior A 2. Neuromodulation: beyond the wiring diagram, M. Higley Wasatch B adding functional flexibility to neural circuits. 3. Sensorimotor processes reflected in spatiotemporal Z. Kilpatrick Superior B of neuronal activity. J-Y. Wu 4. Characterizing neural responses to structured K. Rajan Wasatch A and naturalistic stimuli. W.Bialek 5. Neurophysiological and computational mechanisms of D. Freedman Magpie A categorization. XJ. Wang 6. Perception and decision making in rodents S. Jaramillio Magpie B A. Zador 7. Is it time for theory in olfaction? V.Murphy Maybird N. Uchida G. Otazu C. Poo Workshop Co-Chairs Email Cell Brent Doiron, Pitt [email protected] 412-576-5237 Jess Cardin, Yale [email protected] 267-235-0462 Maps of Snowbird are at the end of this booklet (page 32). 1 COSYNE 2012 Workshops February 27 & 28, 2012 Snowbird, Utah Tuesday, February 28 Organizer(s) Location 1. Understanding heterogeneous cortical activity: S. Ardid Wasatch A the quest for structure and randomness. A. Bernacchia T. Engel 2. Humans, neurons, and machines: how can N. Majaj Wasatch B psychophysics, physiology, and modeling collaborate E. Issa to ask better questions in biological vision J. DiCarlo 3. Inhibitory synaptic plasticity T. Vogels Magpie A H Sprekeler R. Fromeke 4. Functions of identified microcircuits A. Hasenstaub Superior B V. Sohal 5. Promise and peril: genetics approaches for systems K. Nagal Superior A neuroscience revisited. D. Schoppik 6. Perception and decision making in rodents S. -
Music from the Archive of Lazare Saminsky
YIVO INSTITUTE FOR JEWISH RESEARCH PRESENTS JEWISH SONGS AND DANCES: Music from the Archive of Lazare Saminsky SIDNEY KRUM YOUNG ARTISTS CONCERT SERIES In Partnership with Temple Emanu-El · 5th Avenue at 65th Street, NYC June 21, 2017 · 7:00pm PROGRAM The Sidney Krum Young Artists Concert Series is made possible by a generous gift from the Estate of Sidney Krum. In partnership with Temple Emanu-El. Saminsky, Hassidic Suite Op. 24 1-3 Saminsky, First Hebrew Song Cycle Op. 12 1-3 Engel, Omrim: Yeshnah erets Achron, 2 Hebrew Pieces Op. 35 No. 2 Saminsky, Second Hebrew Song Cycle Op. 13 1-3 Saminsky, A Kleyne Rapsodi Saminsky, And Rabbi Eliezer Said Engel, Rabbi Levi-Yitzkah’s Kaddish Achron, Sher Op. 42 Saminsky, Lid fun esterke Saminsky, Shir Hashirim Streicher, Shir Hashirim Engel, Freylekhs, Op. 21 Performers: Mo Glazman, Voice Eliza Bagg, Voice Brigid Coleridge, Violin Julian Schwartz, Cello Marika Bournaki, Piano COMPOSER BIOGRAPHIES Born in Vale-Gotzulovo, Ukraine in 1882, LAZARE SAMINSKY was one of the founding members of the Society for Jewish Folk Music in St. Petersburg – a group of composers committed to forging a new national style of Jewish classical music infused with Jewish folk melodies and liturgical music. Saminksy’s teachers at the St. Petersburg conservatory included Rimsky-Korsakov and Liadov. Fascinated with Jewish culture, Saminsky went on trips to Southern Russia and Georgia to gather Jewish folk music and ancient religious chants. In the 1910s Saminsky spent a substantial amount of time travelling giving lectures and conducting concerts. Saminsky’s travels brought him to Turkey, Syria, Palestine, Paris, and England. -
Bayesian Modeling of Behavior Fall 2017 Wei Ji Ma
Bayesian modeling of behavior Fall 2017 Wei Ji Ma This syllabus is subject to change. Changes will be announced in class and by email. Description Bayesian inference is the mathematical framework for making optimal decisions and actions when the state of the world is not exactly known. This course will provide an intuitive yet mathematically rigorous introduction to Bayesian models of behavior in perception, memory, decision-making, and cognitive reasoning. While this is primarily a psychology course, we will also discuss connections to economics and neuroscience. This course is not about Bayesian data analysis, but about theories that the brain itself is a Bayesian decision-maker. Nevertheless, we will spend some time on model fitting and model comparison. Prerequisites • Strong command of Calculus 1 or equivalent • Introductory course in probability or probability-based statistics. • Ability to code in Matlab. If you have not coded in Matlab before, you will be ok if you have other programming experience and do a tutorial before the course starts. Email Wei Ji if you have any questions about prerequisites. Lecturer Prof. Wei Ji Ma, [email protected], 212 992 6530 Weekly schedule Lecture Wednesdays, 4-6 pm Meyer 815 Recitation Thursdays, 2-4 pm Meyer 815 Office hours By appointment Meyer 754 (Wei Ji’s office) Materials • Bayesian modeling of perception, by Ma, Kording, and Goldreich. Will be distributed in electronic form. • You will need Matlab. If you have a laptop, please install Matlab on it before the course starts. Instructions for if you are on the Meyer building network: http://localweb.cns.nyu.edu/unixadmin/#august10-2015. -
Bayesian Modeling of Behavior Fall 2018 Wei Ji Ma
Bayesian modeling of behavior Fall 2018 Wei Ji Ma This syllabus is subject to change. Changes will be announced in class and by email. Description Bayesian inference is the mathematical framework for making optimal decisions and actions when the state of the world is not exactly known. This course will provide an intuitive yet mathematically rigorous introduction to Bayesian models of behavior in perception, memory, decision-making, and cognitive reasoning. While this is primarily a psychology course, we will also discuss connections to economics and neuroscience. This course is not about Bayesian data analysis, but about theories that the brain itself is a Bayesian decision-maker. Nevertheless, we will spend some time on model fitting and model comparison. Prerequisites • Strong command of Calculus 1 or equivalent • Introductory course in probability or probability-based statistics. • Ability to code in Matlab. If you have not coded in Matlab before, you will be ok if you have other programming experience and do a tutorial before the course starts. Email Wei Ji if you have any questions about prerequisites. Lecturer Prof. Wei Ji Ma, [email protected], 212 992 6530 Weekly schedule Lecture Wednesdays, 4-5:50 pm Meyer 851 Recitation Thursdays, 4-5:50 pm Meyer 851 Office hours By appointment Meyer 754 (Wei Ji’s office) Materials • Bayesian modeling of perception, by Ma, Kording, and Goldreich. Will be distributed in electronic form. • You will need Matlab. Instructions for installing Matlab if you are on the Meyer building network: http://localweb.cns.nyu.edu/unixadmin/#august10-2015. 1 Grading The total grade will be calculated as follows: Best 8 of 11 homework sets 55% Project 30% Participation 15% Letter grade Your numerical grade will be turned into a letter grade according to the following scale: 90-100 A; 87-89 A-; 84-86 B+; 80-83 B; 77-79 B-; 74-76 C+; 70-73 C; 67-69 C-; 64-66 D+; 60-63 D; 57-59 D-; 0-56 F. -
Testing the Bayesian Brain Hypothesis in Visual Perception
TESTING THE BAYESIAN BRAIN HYPOTHESIS IN VISUAL PERCEPTION A Dissertation presented to the Faculty of the Graduate School at the University of Missouri In Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy by APRIL SWAGMAN Dr. Jeffrey N. Rouder, Dissertation Supervisor JULY 2016 The undersigned, appointed by the Dean of the Graduate School, have examined the dissertation entitled: TESTING THE BAYESIAN BRAIN HYPOTHESIS IN VISUAL PERCEPTION presented by April Swagman, a candidate for the degree of Doctor of Philosophy and hereby certify that, in their opinion, it is worthy of acceptance. Dr. Jeffrey N. Rouder Dr. Clint Davis-Stober Dr. Chi-Ren Shyu Dr. Jeffrey Johnson ACKNOWLEDGMENTS For providing the inspiration for this project and my first experiment, I would like to acknowledge Dr. Wei Ji Ma at New York University, who was sure we would find evidence that people are ideally Bayesian. To all past and present members of the Perception and Cognition Lab{Jon Thiele, Julia Haaf, Jory Province, and Tim Ricker{thank you for the encouragement and friendship you provided as I worked on this project. I would especially like to thank Jory for being a mentor and friend throughout my learning experience in the Rouder lab, and for the significant theoret- ical contributions he made to this work. I would like to thank all the PCL undergrad- uate research assistants who helped out with data collection, and especially to Ryan Bahr who helped me become a more effective mentor and was even willing to learn R along the way. And of course I am eternally grateful to my adviser Jeff Rouder for being helpful, inspiring, and, most importantly, patient as I made my way through this work.