alexander-turner / attainable-utility-preservation
☆12Updated 3 years ago
Related projects: ⓘ
- Code for reproducing the results from the paper Avoiding Side Effects in Complex Environments☆12Updated 3 years ago
- ☆9Updated 4 years ago
- An environment for benchmarking commonsense agents☆28Updated 4 years ago
- SafeLife: safety benchmarks for reinforcement learning agents☆59Updated 3 years ago
- ☆80Updated 11 months ago
- Train self-modifying neural networks with neuromodulated plasticity☆77Updated 4 years ago
- PAIRED in PyTorch 🔥☆56Updated last year
- This repository contains code for the method and experiments of the paper "Learning with AMIGo: Adversarially Motivated Intrinsic Goals".☆61Updated last year
- ☆20Updated 5 years ago
- Library to compare and evaluate reward functions☆61Updated 10 months ago
- Reward Learning by Simulating the Past☆43Updated 5 years ago
- ☆10Updated last year
- ☆85Updated 3 years ago
- Experiments in applying interpretability techniques to learned reward functions.☆9Updated 3 years ago
- Library that provides environments for planning problems☆13Updated 2 weeks ago
- Adapting the AlphaZero algorithm to remove the need of execution traces to train NPI.☆77Updated 11 months ago
- Map-Elites based on Evolution Strategies☆31Updated 2 years ago
- Code for "Learning Compositional Rules via Neural Program Synthesis"☆58Updated 3 years ago
- The first place solution for the NeurIPS 2021 Nethack Challenge -- https://www.aicrowd.com/challenges/neurips-2021-the-nethack-challenge☆54Updated last year
- Moore Machine Networks (MMN): Learning Finite-State Representations of Recurrent Policy Networks☆47Updated last year
- Redwood Research's transformer interpretability tools☆11Updated 2 years ago
- A framework for experimenting with never-ending learning☆72Updated 2 years ago
- Interpretability dashboard for reinforcement learners☆16Updated 5 years ago
- ☆29Updated 2 years ago
- Language-annotated Abstraction and Reasoning Corpus☆76Updated last year
- Code for "Meta Learning Backpropagation And Improving It" @ NeurIPS 2021 https://arxiv.org/abs/2012.14905☆31Updated 2 years ago
- Plannable Approximations to MDP Homomorphisms: Equivariance under Actions☆26Updated 4 years ago
- Clockwork VAEs in JAX/Flax☆31Updated 3 years ago
- ☆14Updated 5 years ago
- ☆24Updated 5 years ago