gym-modeestimation

Gym environment class and examples for Mode Estimation Task.

Links

• Original Paper: Akshat Agarwal, Abhinau Kumar V, Kyle Dunovan, Erik Peterson, Timothy Verstynen, Katia Sycara. 2018. Better Safe than Sorry: Evidence Accumulation Allows for Safe Reinforcement Learning. arXiv preprint arXiv:1809.09147. https://arxiv.org/abs/1809.09147

• OpenAI Gym https://github.com/openai/gym

• Beta Distribution https://www.slideshare.net/matsukenbook/6-lt-59735455

Install

pip install gym

git clone https://github.com/susumuota/gym-modeestimation.git
cd gym-modeestimation
pip install -e .
cd ..

Uninstall

pip uninstall gym-modeestimation
pip uninstall gym

Example

simplest example

import gym
import gym.spaces
import gym_modeestimation

def main():
    env = gym.make('ModeEstimation0-v0')
    #env = gym.make('ModeEstimationOnehot0-v0')
    #env = gym.make('ModeEstimationBinary0-v0')
    obs = env.reset()
    for i in range(40):
        action = env.action_space.sample()
        #action = env.NOOP
        obs, reward, done, info = env.step(action)
        print(env.t, env.n0, action, obs, reward, done, info)
        if done:
            obs = env.reset()
    env.close()

if __name__ == '__main__':
    main()

Available Environments

ModeEstimationEPS00-v0
ModeEstimationEPS02-v0
ModeEstimationEPS04-v0
ModeEstimationEPS06-v0
ModeEstimationEPS08-v0
ModeEstimationEPS10-v0

See https://github.com/susumuota/gym-modeestimation/blob/master/gym_modeestimation/__init__.py for more details.

You can add custom environments. See these pages.

https://github.com/openai/gym/blob/master/gym/envs/README.md https://github.com/openai/gym/blob/master/gym/envs/__init__.py

Author

Susumu OTA susumu dot ota at g mail dot com