csp.yaml

name: csp
seed: 0

defaults:
  - logger: wandb
  - scenario: halfcheetah/forgetting
  - hydra: hydra
  - framework/params/algorithm/generic_sac
  - framework/params/policy_agent/generic_policy
  - framework/params/critic_agent/generic_critic
  - framework/params/evaluation/generic_evaluation
  - framework/params/alpha_search/generic_alpha_search
  - optional framework/params/evaluation: ${scenario}
  - optional framework/params/algorithm: ${scenario}
  - optional framework/params/policy_agent: csp/${scenario}
  - optional framework/params/critic_agent: ${scenario}
  - optional framework/params/alpha_search: ${scenario}

framework:
  classname: frameworks.Subspace
  seed: ${seed}
  params:
    checkpoint: False

    evaluation:
      oracle_rollouts: 0

    lr_scaling: 0. #scaling of the lr * nb_anchors

    policy_agent:
      classname: agents.SubspaceActionAgent
      n_initial_anchors: 1
#      dist_type: flat
      refresh_rate: 1.
      resampling_q: True
      resampling_policy: True
#      repeat_alpha: 100

    critic_agent:
      classname: agents.AlphaTwinCritics
      n_anchors: