Random Agent

class or_suite.agents.rl.random.randomAgent

Randomized RL Algorithm

Implements the randomized RL algorithm - selection an action uniformly at random from the action space. In particular, the algorithm stores an internal copy of the environment’s action space and samples uniformly at random from it.

__init__()

Initialize self. See help(type(self)) for accurate signature.

pick_action(obs, h)

Selects an action for the algorithm.

Parameters

• obs – a state for the environment

• h – timestep

Returns

An action sampled uniformly at random from the environment’s action space.

update_config(env, config=None)

Updates configuration file for the agent

Updates the stored environment to sample uniformly from.

Parameters

• env – an openAI gym environment

• config – an (optional) dictionary containing parameters for the environment

update_obs(obs, action, reward, newObs, timestep, info)

Add observation to records

update_policy(h)

Update internal policy based upon records