Obserwuj
Ted Moskovitz
Ted Moskovitz
Anthropic
Zweryfikowany adres z anthropic.com - Strona główna
Tytuł
Cytowane przez
Cytowane przez
Rok
Feedback alignment in deep convolutional networks
TH Moskovitz, A Litwin-Kumar, LF Abbott
arXiv preprint arXiv:1812.06488, 2018
752018
Tactical Optimism and Pessimism for Deep Reinforcement Learning
T Moskovitz, J Parker-Holder, A Pacchiano, M Arbel, MI Jordan
Neural Information Processing Systems (NeurIPS) 2021, 2021
662021
Confronting Reward Model Overoptimization with Constrained RLHF
T Moskovitz, AK Singh, DJ Strouse, T Sandholm, R Salakhutdinov, ...
International Conference on Learning Representations (ICLR) 2024, 2023
412023
Action prediction error: a value-free dopaminergic teaching signal that drives stable learning
F Greenstreet, HM Vergara, S Pati, L Schwarz, M Wisdom, F Marbach, ...
bioRxiv, 2022
372022
The Transient Nature of Emergent In-context Learning in Transformers
AK Singh*, SCY Chan*, T Moskovitz, A Saxe, F Hill
NeurIPS 2023, 2023
332023
Efficient Wasserstein Natural Gradients for Reinforcement Learning
T Moskovitz*, M Arbel*, F Huszar, A Gretton
International Conference on Learning Representations (ICLR) 2021, 2020
272020
ReLOAD: Reinforcement Learning with Optimistic Ascent-Descent for Last-Iterate Convergence in Constrained MDPs
T Moskovitz, B O’Donoghue, V Veeriah, S Flennerhag, S Singh, T Zahavy
International Conference on Machine Learning, 25303-25336, 2023
232023
What needs to go right for an induction head? a mechanistic study of in-context learning circuits and their formation
AK Singh, T Moskovitz, F Hill, SCY Chan, AM Saxe
arXiv preprint arXiv:2404.07129, 2024
202024
A First-Occupancy Representation for Reinforcement Learning
T Moskovitz, SR Wilson, M Sahani
International Conference on Learning Representations (ICLR) 2022, 2021
162021
A comparison of deep learning and linear-nonlinear cascade approaches to neural encoding
TH Moskovitz, NA Roy, JW Pillow
BioRxiv, 463422, 2018
122018
Towards an Understanding of Default Policies in Multitask Policy Optimization
T Moskovitz, M Arbel, J Parker-Holder, A Pacchiano
Conference on Artificial Intelligence and Statistics (AISTATS) 2022, 2021
102021
First-Order Preconditioning via Hypergradient Descent
T Moskovitz, R Wang, J Lan, S Kapoor, T Miconi, J Yosinski, A Rawal
NeurIPS 2019, Beyond First Order Methods in ML Workshop, 2019
8*2019
Understanding dual process cognition via the minimum description length principle
T Moskovitz, KJ Miller, M Sahani, MM Botvinick
PLOS Computational Biology 20 (10), e1012383, 2024
7*2024
Understanding the functional and structural differences across excitatory and inhibitory neurons
S Minni, L Ji-An, T Moskovitz, G Lindsay, K Miller, M Dipoppa, GR Yang
bioRxiv, 680439, 2019
52019
Minimum Description Length Control
T Moskovitz, TC Kao, M Sahani, MM Botvinick
International Conference on Learning Representations (ICLR) 2023, 2022
32022
Amortised Learning by Wake-Sleep
LK Wenliang, T Moskovitz, H Kanagawa, M Sahani
International Conference on Machine Learning (ICML) 2020, 2020
32020
HARP: A challenging human-annotated math reasoning benchmark
AS Yue, L Madaan, T Moskovitz, DJ Strouse, AK Singh
arXiv preprint arXiv:2412.08819, 2024
12024
A State Representation for Diminishing Rewards
T Moskovitz, S Hromadka, A Touati, D Borsa, M Sahani
Thirty-seventh Conference on Neural Information Processing Systems, 2023
12023
Do Biologically-Realistic Recurrent Architectures Produce Biologically-Realistic Models?
G Lindsay, T Moskovitz, GR Yang, K Miller
Conference on Cognitive Computational Neuroscience (CCN) 2019, 2019
12019
Nie można teraz wykonać tej operacji. Spróbuj ponownie później.
Prace 1–19