tmaze#
- hybrid_aif.reference.envs.tmaze(
- reward_probability=1.0,
- punishment_probability=1.0,
- cue_validity=0.95,
- reward_condition=None,
The classic TMaze as a positional process and its parameters, tensor-identical to pymdp’s.
Two state factors (location, five levels, controlled; context, two levels, fixed) and three modalities (location, reward, cue; see the module docstring).
reward_probabilityis the chance the rewarded arm pays,punishment_probabilitythe chance the other arm punishes,cue_validitythe chance the cue location shows the true context, andreward_conditionfixes the context (None: 50/50). The probabilities may be traced, so batches of parameter sets build undervmap.