Ted Moskovitz

Publication Activity (10 Years)

Years Active: 2019-2024
Publications (10 Years): 21

Top Topics

Minimum Description Length

Factored Markov Decision Processes

Reinforcement Learning

Top Venues

Publications

Ted Moskovitz, Aaditya K. Singh, DJ Strouse, Tuomas Sandholm, Ruslan Salakhutdinov, Anca D. Dragan, Stephen Marcus McAleer
Confronting Reward Model Overoptimization with Constrained RLHF. ICLR (2024)
Aaditya K. Singh, Ted Moskovitz, Felix Hill, Stephanie C. Y. Chan, Andrew M. Saxe
What needs to go right for an induction head? A mechanistic study of in-context learning circuits and their formation. CoRR (2024)
Ted Moskovitz, Aaditya K. Singh, DJ Strouse, Tuomas Sandholm, Ruslan Salakhutdinov, Anca D. Dragan, Stephen McAleer
Confronting Reward Model Overoptimization with Constrained RLHF. CoRR (2023)
Aaditya K. Singh, Stephanie C. Y. Chan, Ted Moskovitz, Erin Grant, Andrew M. Saxe, Felix Hill
The Transient Nature of Emergent In-Context Learning in Transformers. NeurIPS (2023)
Ted Moskovitz, Ta-Chu Kao, Maneesh Sahani, Matt M. Botvinick
Minimum Description Length Control. ICLR (2023)
Aaditya K. Singh, Stephanie C. Y. Chan, Ted Moskovitz, Erin Grant, Andrew M. Saxe, Felix Hill
The Transient Nature of Emergent In-Context Learning in Transformers. CoRR (2023)
Ted Moskovitz, Brendan O'Donoghue, Vivek Veeriah, Sebastian Flennerhag, Satinder Singh, Tom Zahavy
ReLOAD: Reinforcement Learning with Optimistic Ascent-Descent for Last-Iterate Convergence in Constrained MDPs. ICML (2023)
Ted Moskovitz, Samo Hromadka, Ahmed Touati, Diana Borsa, Maneesh Sahani
A State Representation for Diminishing Rewards. NeurIPS (2023)
Ted Moskovitz, Brendan O'Donoghue, Vivek Veeriah, Sebastian Flennerhag, Satinder Singh, Tom Zahavy
ReLOAD: Reinforcement Learning with Optimistic Ascent-Descent for Last-Iterate Convergence in Constrained MDPs. CoRR (2023)
Ted Moskovitz, Samo Hromadka, Ahmed Touati, Diana Borsa, Maneesh Sahani
A State Representation for Diminishing Rewards. CoRR (2023)
Ted Moskovitz, Spencer R. Wilson, Maneesh Sahani
A First-Occupancy Representation for Reinforcement Learning. ICLR (2022)
Abhi Gupta, Ted Moskovitz, David Alvarez-Melis, Aldo Pacchiano
Transfer RL via the Undo Maps Formalism. CoRR (2022)
Ted Moskovitz, Ta-Chu Kao, Maneesh Sahani, Matthew M. Botvinick
Minimum Description Length Control. CoRR (2022)
Ted Moskovitz, Michael Arbel, Jack Parker-Holder, Aldo Pacchiano
Towards an Understanding of Default Policies in Multitask Policy Optimization. AISTATS (2022)
Ted Moskovitz, Spencer R. Wilson, Maneesh Sahani
A First-Occupancy Representation for Reinforcement Learning. CoRR (2021)
Ted Moskovitz, Michael Arbel, Jack Parker-Holder, Aldo Pacchiano
Towards an Understanding of Default Policies in Multitask Policy Optimization. CoRR (2021)
Ted Moskovitz, Michael Arbel, Ferenc Huszar, Arthur Gretton
Efficient Wasserstein Natural Gradients for Reinforcement Learning. ICLR (2021)
Ted Moskovitz, Jack Parker-Holder, Aldo Pacchiano, Michael Arbel, Michael I. Jordan
Tactical Optimism and Pessimism for Deep Reinforcement Learning. NeurIPS (2021)
Ted Moskovitz, Jack Parker-Holder, Aldo Pacchiano, Michael Arbel
Deep Reinforcement Learning with Dynamic Optimism. CoRR (2021)
Ted Moskovitz, Michael Arbel, Ferenc Huszar, Arthur Gretton
Efficient Wasserstein Natural Gradients for Reinforcement Learning. CoRR (2020)
Ted Moskovitz, Rui Wang, Janice Lan, Sanyam Kapoor, Thomas Miconi, Jason Yosinski, Aditya Rawal
First-Order Preconditioning via Hypergradient Descent. CoRR (2019)