research
∙
09/05/2023
Explaining grokking through circuit efficiency
One of the most surprising puzzles in neural network generalisation is g...
share
research
∙
07/28/2023
The Hydra Effect: Emergent Self-repair in Language Model Computations
We investigate the internal structure of language model computations usi...
research
∙
07/18/2023
Does Circuit Analysis Interpretability Scale? Evidence from Multiple Choice Capabilities in Chinchilla
Circuit analysis is a promising technique for understanding the internal...
research
∙
04/13/2023
Power-seeking can be probable and predictive for trained agents
Power-seeking behavior is a key source of risk from advanced AI, but our...
research
∙
01/12/2023
Tracr: Compiled Transformers as a Laboratory for Interpretability
Interpretability research aims to build tools for understanding machine ...
research
∙
06/02/2021
Sample-based Approximation of Nash in Large Many-Player Games via Gradient Descent
Nash equilibrium is a central concept in game theory. Several Nash solve...
research
∙
06/08/2020
Learning to Play No-Press Diplomacy with Best Response Policy Iteration
Recent advances in deep reinforcement learning (RL) have led to consider...
research
∙
04/16/2020
Should I tear down this wall? Optimizing social metrics by evaluating novel actions
One of the fundamental challenges of governance is deciding when and how...
research
∙
08/26/2019
OpenSpiel: A Framework for Reinforcement Learning in Games
OpenSpiel is a collection of environments and algorithms for research in...
research
∙
03/19/2019
Learning Reciprocity in Complex Sequential Social Dilemmas
Reciprocity is an important feature of human social interaction and unde...
research
∙
02/26/2018
Reinforcement and Imitation Learning for Diverse Visuomotor Skills
We propose a model-free deep reinforcement learning method that leverage...
research
∙
07/24/2017
Guidelines for Artificial Intelligence Containment
With almost daily improvements in capabilities of artificial intelligenc...
research
∙
06/03/2016
Zoneout: Regularizing RNNs by Randomly Preserving Hidden Activations
We propose zoneout, a novel method for regularizing RNNs. At each timest...
research
∙
04/02/2016