diff --git a/README.md b/README.md index 444206c..e8486c6 100644 --- a/README.md +++ b/README.md @@ -5,7 +5,7 @@ -**Status**: alpha, pre-release. +**Status**: early beta. *seals*, the Suite of Environments for Algorithms that Learn Specifications, is a toolkit for evaluating specification learning algorithms, such as reward or imitation learning. The @@ -13,13 +13,14 @@ environments are compatible with [Gym](https://github.com/openai/gym), but are d to test algorithms that learn from user data, without requiring a procedurally specified reward function. -This is currently a work-in progress. While you are welcome to use the repository, we may make -breaking changes at any time without prior notice. We intend for it to eventually contain: +There are two types of environments in *seals*: - - A suite of diagnostic tasks for reward learning. - - Wrappers around common RL benchmark environments that help to avoid common pitfalls in - benchmarking (e.g. by masking visible score counters in Gym Atari tasks). - - New challenge tasks for specification learning algorithms. + - **Diagnostic Tasks** which test individual facets of algorithm performance in isolation. + - **Renovated Environments**, adaptations of widely-used benchmarks such as MuJoCo continuous + control tasks to be suitable for specification learning benchmarks. In particular, we remove + any side-channel sources of reward information. + +*seals* is under active development and we intend to add more categories of tasks soon. You may also be interested in our sister project [imitation](https://github.com/humancompatibleai/imitation/), providing implementations of a variety of imitation and reward learning algorithms.