Categories
Kreiman Lab News

Statistical Learning Theory and Applications

MIT Class 9.520 / 6.860 (Fall 2021)

Class 19 and Class 20: Real neural networks and the ventral stream

By Thomas Serre, Gabriel Kreiman and Tomaso Poggio

Slides Nov 16, 2021

Categories
Kreiman Lab News

Shashi Gupta and Mengmi Zhang present their work at NeurIPS

Gupta SK, Zhang M, Wu CC, Wolfe JM, Kreiman G (2021). Visual Search Asymmetry: Deep Nets and Humans Share Similar Inherent Biases. NeurIPS arXiv 2106.02953 PDF | Supplementary Material | Resources

Gupta et al NeurIPS 2021
Eccentricity-dependent sampling in the visual system
Categories
Kreiman Lab News

Trenton Bricken presents his work at NeurIPS

SDM
Relationship between sparse distributed memory and transformer architectures

Attention approximates sparse distributed memory

Bricken T and Pehlevan C. NeurIPS 2021

Categories
Kreiman Lab News

Introduction to computational and theoretical neuroscience

November 9, 2021

Link to slides

Categories
Kreiman Lab News

CBMM: The brain’s operating system. Research Update 2

We hope that you will be able to join next week’s research meeting with presentations by Mengmi Zhang and Jie Zheng, Kreiman Lab.

CBMM Research Meeting

Title: Module 2 Research presentation

Date/Time: October 26, 2021, 4:00 pm to 5:30 pm ET

Jie Zheng
Jie Zheng

Jie Zheng‘s presentation (in person):

Title: Neurons that structure memories of ordered experience in human

 Abstract: The process of constructing temporal associations among related events is essential to episodic memory. However, what neural mechanism helps accomplish this function remains unclear. To address this question, we recorded single unit activity in humans while subjects performed a temporal order memory task. During encoding, subjects watched a series of clips (i.e., each clip consisted of 4 events) and were later instructed to retrieve the ordinal information of event sequences. We found that hippocampal neurons in humans could index specific orders of events with increased neuronal firings (i.e., rate order cells) or clustered spike timing relative to theta phases (i.e., phase order cells), which are transferrable across different encoding experiences (e.g., different clips). Rate order cells also increased their firing rates when subjects correctly retrieved the temporal information of their preferred ordered events. Phase order cells demonstrated stronger phase precessions at event transitions during encoding for clips whose ordinal information was subsequently correct retrieved. These results not only highlight the critical role of the hippocampus in structuring memories of continuous event sequences but also suggest a potential neural code representing temporal associations among events.

Mengmi Zhang
Mengmi Zhang

Mengmi Zhang‘s [virtual] presentation:

Title: Visual Search Asymmetry: Deep Nets and Humans Share Similar Inherent Biases

Abstract: Visual search is a ubiquitous and often challenging daily task, exemplified by looking for the car keys at home or a friend in a crowd. An intriguing property of some classical search tasks is an asymmetry such that finding a target A among distractors B can be easier than finding B among A. To elucidate the mechanisms responsible for asymmetry in visual search, we propose a computational model that takes a target and a search image as inputs and produces a sequence of eye movements until the target is found. The model integrates eccentricity-dependent visual recognition with target-dependent top-down cues. We compared the model against human behavior in six paradigmatic search tasks that show asymmetry in humans. Without prior exposure to the stimuli or task-specific training, the model provides a plausible mechanism for search asymmetry. We hypothesized that the polarity of search asymmetry arises from experience with the natural environment. We tested this hypothesis by training the model on an augmented version of ImageNet where the biases of natural images were either removed or reversed. The polarity of search asymmetry disappeared or was altered depending on the training protocol. This study highlights how classical perceptual properties can emerge in neural network models, without the need for task-specific training, but rather as a consequence of the statistical properties of the developmental diet fed to the model. Our work will be presented in the upcoming Neurips conference, 2021.

See also: Gupta SK, Zhang M, Wu CC, Wolfe JM, Kreiman G (2021). Visual search asymmetry: deep nets and humans share similar inherent biases. NeurIPS PDF