Sreejan Kumar
Sreejan Kumar
PhD Candidate, Princeton University
Потвърден имейл адрес: princeton.edu - Начална страница
Shared functional specialization in transformer-based language models and the human brain
S Kumar, TR Sumers, T Yamakoshi, A Goldstein, U Hasson, KA Norman, ...
Nature Communications, 2024
Using Natural Language and Program Abstractions to Instill Human Inductive Biases in Machines
S Kumar, CG Correa, I Dasgupta, R Marjieh, MY Hu, RD Hawkins, ...
Neural Information Processing Systems (NeurIPS), 2022
Meta-Learning of Structured Task Distributions in Humans and Machines
S Kumar, I Dasgupta, JD Cohen, ND Daw, TL Griffiths
International Conference of Learning Representations (ICLR) 2021, 2021
An information network flow approach for measuring functional connectivity and predicting behavior
S Kumar, K Yoo, MD Rosenberg, D Scheinost, RT Constable, S Zhang, ...
Brain and behavior 9 (8), e01346, 2019
Disentangling abstraction from statistical pattern matching in human and machine learning
S Kumar, I Dasgupta, ND Daw, JD Cohen, TL Griffiths
PLoS computational biology 19 (8), e1011316, 2023
Searching through functional space reveals distributed visual, auditory, and semantic coding in the human brain
S Kumar, CT Ellis, TP O’Connell, MM Chun, NB Turk-Browne
PLOS Computational Biology 16 (12), e1008457, 2020
Relational Constraints On Neural Networks Reproduce Human Biases towards Abstract Geometric Regularity
D Campbell, S Kumar, T Giallanza, JD Cohen, TL Griffiths
arXiv preprint arXiv:2309.17363, 2023
On the hazards of relating representations and inductive biases
TL Griffiths, S Kumar, RT McCoy
Behavioral and Brain Sciences 46, e275, 2023
Human-Like Geometric Abstraction in Large Pre-trained Neural Networks
D Campbell, S Kumar, T Giallanza, TL Griffiths, JD Cohen
arXiv preprint arXiv:2402.04203, 2024
Comparing Abstraction in Humans and Large Language Models Using Multimodal Serial Reproduction
S Kumar, R Marjieh, B Zhang, D Campbell, MY Hu, U Bhatt, B Lake, ...
arXiv preprint arXiv:2402.03618, 2024
Learning to Abstract Visuomotor Mappings using Meta-Reinforcement Learning
CA Velazquez-Vargas, IR Christian, JA Taylor, S Kumar
arXiv preprint arXiv:2402.03072, 2024
Using Natural Language to Guide Meta-Learning Agents towards Human-like Inductive Biases
S Kumar, I Dasgupta, M Hu, R Marjieh, RD Hawkins, N Daw, J Cohen, ...
ACL Workshop on Learning with Natural Language Supervision, 2022
Системата не може да изпълни операцията сега. Опитайте отново по-късно.
Статии 1–12