Tuesday, October 29, 2013 — 3:30 PM to 5:00 PM EDT

Surya Ganguli
Department of Applied Physics
Stanford University

A Theory of Neural Dimensionality and Dynamics

In a wide variety of experimental paradigms, neuroscientists tightly control behavior, record many trials, and obtain trial averaged neuronal firing rate data from hundreds of neurons, in circuits containing millions to billions of behaviorally relevant neurons. Such datasets are often analyzed by dimensionality reduction methods that allow us to visualize neuronal dynamics through their projections onto a number of basis patterns. Strikingly, recordings from hundreds of neurons can often be described using a much smaller number of dimensions (basis patterns), and the resulting projections yield a remarkably insightful dynamical portrait of neural circuit computation. Thus many neuronal datasets are surprisingly simple, and we seem to be able to extract reasonable collective neuronal dynamics despite overwhelming levels of neuronal subsampling. This ubiquitous simplicity raises several profound and timely conceptual questions. What is the origin of this simplicity? What does it tell us about the complexity of brain dynamics? Would neuronal datasets become more complex if we recorded more neurons? How and when can we trust dynamical portraits obtained from only hundreds of neurons in a circuit containing billions of neurons? More generally, what, if anything, can we learn about a complex dynamical system by measuring an infinitesimal fraction of its degrees of freedom? We present a theory of neural dimensionality and dynamics that answers all of these questions, and we further test this theory in neural recordings from monkeys performing reaching movements.

Location 
PAS - Psychology, Anthropology, Sociology
Room 2464
200 University Avenue West

Waterloo, ON N2L 3G1
Canada

S M T W T F S
25
26
27
28
29
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
1
2
3
4
5
6
  1. 2024 (4)
    1. April (1)
    2. March (2)
    3. February (1)
  2. 2023 (9)
    1. December (1)
    2. November (1)
    3. October (2)
    4. September (1)
    5. April (1)
    6. March (1)
    7. February (1)
    8. January (1)
  3. 2022 (6)
  4. 2021 (3)
  5. 2020 (4)
  6. 2019 (7)
  7. 2018 (4)
  8. 2017 (7)
  9. 2016 (8)
  10. 2015 (9)
  11. 2014 (6)
  12. 2013 (8)
  13. 2012 (4)

Brain Day 2023 Videos On-line

The videos from Brain Day 2023 are now available on line at our youtube channel. Hope you enjoy.

CTN Masters Student Graduate Sugandha Sharma Appears on Generally Intelligent Podcast

Sugandha Sharma, masters student graduate of the University of Waterloo's CTN, discusses her research and time in the laboratory of CTN Founding Director Chris Eliasmith as well as her current PhD research at MIT on the Generally Intelligent Podcast. Give it a listen.

Sue Ann Campbell Presents at International Conference on Mathematical Neurosci 2022

Sue Ann Campbell (Applied Math/CTN core member) recently presented "Modulation of Synchronization by a Slowly Varying Current"  in July 2022 at the International Conference on Mathematical Neuroscience; Watch it on YouTubesue ann campbell presentation image of spikes

CTN Research Day 2023 Oct 17 16:30 - 19:00 QNC 0101

The Centre for Theoretical Neuroscience will be hosting its second Research Day. This will be a chance to start the new academic year by getting re-acquainted with each other and the diversity of research conducted by CTN core and affiliate faculty. The format will be to have a number of CTN faculty share short overviews of their lab's and projects (16:30-17:30) and then, following a short coffee break (17:30-18:00), hear from a dozen current graduate students and post-docs giving short three minute talks on an aspect of their current research (18:00-19:00).

Bots and Beasts. New book by CTN Founding Member Paul Thagard

Paul Thagard, philosopher, cognitive scientist, Killam prize winner, and founding CTN member has a new book out: Bots and Beasts. bots and beasts book cover