Datasets in neuroscience are increasing in size at alarming rates relative to our ability to analyze them. This workshop aims at discussing new frameworks for processing and making sense of large neural datasets. The morning session will focus on approaches for processing large neuroscience datasets. Examples include: distributed + high-performance computing, GPU and other hardware accelerations, spatial databases, and other compression schemes used for large neuroimaging datasets, online machine learning approaches for handling large data sizes, randomization and stochastic optimization. The afternoon session will focus on abstractions for modelling large neuroscience datasets. Examples include graphs, graphical models, manifolds, mixture models, latent variable models, spatial models, and factor learning.
Have a question that isn't addressed below? Email us at bigneurodata@gmail.com
The morning will consist of two sessions focused on approaches for processing large neuroscience datasets. There will be three or four invited talks and one submitted talk per session, as well as time for an open panel discussion.
This format will be repeated in the afternoon, alongside a poster session and spotlight, as well as an open discussion at the end of the day.
We welcome participants from the machine learning community, data scientists, and neuroscientists alike!
If you don't know about neuroscience, don’t worry - we will introduce the basics behind some promising big neuro datasets and their associated computational challenges! This will set the stage for the afternoon talks and discussion, where we can talk about how machine learning can aid in analyzing these datasets.
The workshop is part of the NIPS workshop program, held at the Long Beach Convention Center, in Long Beach, CA. The event will take place from 8:30-5:30 on Saturday Dec 9th, 2017.
Both the morning and afternoon will be broken up into two sessions consisting of talks and a round-table discussion of the speakers. The afternoon will also consist of poster presentations.
Submit your contribution for a talk or poster here!
The submission deadline is October 31st, 2017, and the decisions will be announced on November 10th, 2017. To be considered for a talk, please include a URL to a pre-print of an unpublished manuscript on arXiv (or related preprint service, such as biorRxiv). To be considered for a poster, please attach a PDF abstract of your proposed topic. Please attach only one of these documents in a single submission.
Welcome & Introductory Remarks
| Time | Presentation | Location |
|---|---|---|
| 8:45-9:00 | Eva Dyer, Georgia Institute of Technology | Room 204 |
Deep Learning & Neuro Session
| Time | Presentation | Location |
|---|---|---|
| 9:00-9:35 | Jim DiCarlo, Massachusetts Institute of Technology, “Can brain data be used to reverse engineer the algorithms of human perception?” |
Room 204 |
| 9:35-9:55 | Timothy Lillicrap, Google DeepMind, “Backpropagation and deep learning in the brain” |
Room 204 |
| 9:55-10:15 | Viren Jain, Google, “Algorithms, tools, and progress in connectomic reconstruction of neural circuits” |
Room 204 |
| 10:15-10:35 | Bing Brunton, University of Washington, “Multimodal deep learning for natural human neural recordings and video” |
Room 204 |
| 10:35-11:00 | Coffee Break | Room 204 |
| 11:00-11:35 | Yoshua Bengio, University of Montreal, "More Steps towards Biologically Plausible Backprop" |
Room 204 |
Panel Discussion
Moderator: Will Gray Roncal
| Time | Presentation | Location |
|---|---|---|
| 11:40-12:20 | "What neural systems can teach us about building better machine learning algorithms?" Panel: Bing Brunton (U Washington), Jim DiCarlo (MIT), Timothy Lillicrap (DeepMind), Viren Jain (Google), Nathan Kutz (U Washington) |
Room 204 |
| 12:20-13:40 | Lunch Break | Room 204 |
Dimensionality Reduction and Adaptive Experiments Session
| Time | Presentation | Location |
|---|---|---|
| 13:40-14:15 | Nathan Kutz, University of Washington, “Discovery of governing equations and biological principles from spatio-temporal time-series recordings” |
Room 204 |
| 14:15-14:35 | Eftychios Pnevmatikakis, Flatiron Institute, Simons Foundation, “Large scale calcium imaging data analysis for the 99%” |
Room 204 |
| 14:35-14:55 | Gael Varoquaux, INRIA “Machine learning for cognitive mapping” |
Room 204 |
Contributed Talks / Poster Spotlights
| Time | Presentation | Location |
|---|---|---|
| 14:55-15:15 | Pierre Bellec / Christian Dansereau, CRIUGM, "Dealing with clinical heterogeneity in the discovery of new biomarkers of brain disorders" |
Room 204 |
| 15:15-15:25 | David Rolnick, MIT, “Morphological error detection for connectomics” |
Room 204 |
| 15:25-15:35 | Poster Spotlights: Shariq Iqbal, Milad Makkie, Sang-Yun Oh | Room 204 |
Poster Session
| Time | Presentation | Location |
|---|---|---|
| 15:45-16:30 | Posters | Room 204 |
| 15:15-15:25 | Coffee Break | Room 204 |
Macroscale Data/Functional Connectivity Session
| Time | Presentation | Location |
|---|---|---|
| 16:30-17:05 | Bin Yu, UC Berkeley, “Deep nets meet real neurons: pattern selectivity of V4 through transfer learning and stability analysis” |
Room 204 |
| 17:05-17:25 | Vince Calhoun, Mind Research Network “Mapping brain structure and function with deep learning” |
Room 204 |
Big Data Session
Moderator: Will Gray Roncal
| Time | Presentation | Location |
|---|---|---|
| 17:25-17:45 | Nathan Drenkow, JHU Applied Physics Lab, “bossDB: A Petascale Database for Large-Scale Neuroscience” |
Room 204 |
| 17:45-18:05 | Kristin Branson, Janelia Farm, "Machine vision and learning for extracting a mechanistic understanding of neural computation" |
Room 204 |
| 18:05-18:30 | Discussion | Room 204 |