Obserwuj
Surbhi Goel
Surbhi Goel
Assistant Professor, University of Pennsylvania
Zweryfikowany adres z cis.upenn.edu - Strona główna
Tytuł
Cytowane przez
Cytowane przez
Rok
Transformers learn shortcuts to automata
B Liu, JT Ash, S Goel, A Krishnamurthy, C Zhang
arXiv preprint arXiv:2210.10749, 2022
1802022
Reliably learning the relu in polynomial time
S Goel, V Kanade, A Klivans, J Thaler
Conference on Learning Theory (COLT) 2017, 2016
1472016
Hidden progress in deep learning: Sgd learns parities near the computational limit
B Barak, B Edelman, S Goel, S Kakade, E Malach, C Zhang
Advances in Neural Information Processing Systems 35, 21750-21764, 2022
1422022
Inductive biases and variable creation in self-attention mechanisms
BL Edelman, S Goel, S Kakade, C Zhang
International Conference on Machine Learning, 5793-5831, 2022
1402022
Understanding contrastive learning requires incorporating inductive biases
N Saunshi, J Ash, S Goel, D Misra, C Zhang, S Arora, S Kakade, ...
International Conference on Machine Learning, 19250-19286, 2022
1292022
Learning neural networks with two nonlinear layers in polynomial time
S Goel, A Klivans
Conference on Learning Theory (COLT) 2019, 2017
113*2017
Learning one convolutional layer with overlapping patches
S Goel, A Klivans, R Meka
International Conference on Machine Learning (ICML) 2018, 2018
892018
Gone fishing: Neural active learning with fisher embeddings
J Ash, S Goel, A Krishnamurthy, S Kakade
Advances in Neural Information Processing Systems 34, 8927-8939, 2021
882021
Superpolynomial lower bounds for learning one-layer neural networks using gradient descent
S Goel, A Gollakota, Z Jin, S Karmalkar, A Klivans
International Conference on Machine Learning, 3587-3596, 2020
842020
Statistical-query lower bounds via functional gradients
S Goel, A Gollakota, A Klivans
Advances in Neural Information Processing Systems 33, 2147-2158, 2020
682020
Approximation schemes for relu regression
I Diakonikolas, S Goel, S Karmalkar, AR Klivans, M Soltanolkotabi
Conference on learning theory, 1452-1485, 2020
612020
Time/accuracy tradeoffs for learning a relu with respect to gaussian marginals
S Goel, S Karmalkar, A Klivans
Advances in neural information processing systems 32, 2019
612019
Investigating the role of negatives in contrastive representation learning
JT Ash, S Goel, A Krishnamurthy, D Misra
arXiv preprint arXiv:2106.09943, 2021
562021
Exposing attention glitches with flip-flop language modeling
B Liu, J Ash, S Goel, A Krishnamurthy, C Zhang
Advances in Neural Information Processing Systems 36, 2023
492023
Tight hardness results for training depth-2 ReLU networks
S Goel, A Klivans, P Manurangsi, D Reichman
arXiv preprint arXiv:2011.13550, 2020
442020
Efficiently learning adversarially robust halfspaces with noise
O Montasser, S Goel, I Diakonikolas, N Srebro
International Conference on Machine Learning, 7010-7021, 2020
362020
The evolution of statistical induction heads: In-context learning markov chains
BL Edelman, E Edelman, S Goel, E Malach, N Tsilivis
arXiv preprint arXiv:2402.11004, 2024
312024
Eigenvalue decay implies polynomial-time learnability for neural networks
S Goel, A Klivans
Advances in Neural Information Processing Systems 30, 2017
302017
Quantifying perceptual distortion of adversarial examples
M Jordan, N Manoj, S Goel, AG Dimakis
arXiv preprint arXiv:1902.08265, 2019
292019
Acceleration via fractal learning rate schedules
N Agarwal, S Goel, C Zhang
International Conference on Machine Learning, 87-99, 2021
282021
Nie można teraz wykonać tej operacji. Spróbuj ponownie później.
Prace 1–20