fqf

fqf#

Source code: tianshou/algorithm/modelfree/fqf.py

class FQFTrainingStats(*, train_time: float = 0.0, smoothed_loss: dict = <factory>, loss: float, quantile_loss: float, fraction_loss: float, entropy_loss: float)[source]#

Bases: SimpleLossTrainingStats

quantile_loss: float#

fraction_loss: float#

entropy_loss: float#

class FQFPolicy(*, model: FullQuantileFunction, fraction_model: FractionProposalNetwork, action_space: Space, observation_space: Space | None = None, eps_training: float = 0.0, eps_inference: float = 0.0)[source]#

Bases: QRDQNPolicy

Parameters:

model – a model following the rules (s_B -> action_values_BA)
fraction_model – a FractionProposalNetwork for proposing fractions/quantiles given state.
action_space – the environment’s action space
observation_space – the environment’s observation space.
eps_training – the epsilon value for epsilon-greedy exploration during training. When collecting data for training, this is the probability of choosing a random action instead of the action chosen by the policy. A value of 0.0 means no exploration (fully greedy) and a value of 1.0 means full exploration (fully random).
eps_inference – the epsilon value for epsilon-greedy exploration during inference, i.e. non-training cases (such as evaluation during test steps). The epsilon value is the probability of choosing a random action instead of the action chosen by the policy. A value of 0.0 means no exploration (fully greedy) and a value of 1.0 means full exploration (fully random).

Compute action over the given batch data.

If you need to mask the action, please add a “mask” into batch.obs, for example, if we have an environment that has “0/1/2” three actions:

batch == Batch(
    obs=Batch(
        obs="original obs, with batch_size=1 for demonstration",
        mask=np.array([[False, True, False]]),
        # action 1 is available
        # action 0 and 2 are unavailable
    ),
    ...
)

Parameters:

batch –
state – optional hidden state (for RNNs)
model – if not passed will use self.model. Typically used to pass the lagged target network instead of using the current model.

Returns:

A Batch which has 3 keys:

act the action.
logits the network’s raw output.
state the hidden state.

class FQF(*, policy: FQFPolicy, optim: OptimizerFactory, fraction_optim: OptimizerFactory, gamma: float = 0.99, num_fractions: int = 32, ent_coef: float = 0.0, n_step_return_horizon: int = 1, target_update_freq: int = 0)[source]#

Bases: QRDQN[FQFPolicy]

Implementation of Fully Parameterized Quantile Function for Distributional Reinforcement Learning. arXiv:1911.02140.

Parameters:

policy – the policy
optim – the optimizer factory for the policy’s main Q-function model
fraction_optim – the optimizer factory for the policy’s fraction model
action_space – the environment’s action space.
gamma – the discount factor in [0, 1] for future rewards. This determines how much future rewards are valued compared to immediate ones. Lower values (closer to 0) make the agent focus on immediate rewards, creating “myopic” behavior. Higher values (closer to 1) make the agent value long-term rewards more, potentially improving performance in tasks where delayed rewards are important but increasing training variance by incorporating more environmental stochasticity. Typically set between 0.9 and 0.99 for most reinforcement learning tasks
num_fractions – the number of fractions to use.
ent_coef – coefficient that weights the entropy bonus relative to the actor loss. Controls the exploration-exploitation trade-off by encouraging policy entropy. Higher values promote more exploration by encouraging a more uniform action distribution. Lower values focus more on exploitation of the current policy’s knowledge. Typically set between 0.01 and 0.05 for most actor-critic implementations.
n_step_return_horizon – the number of future steps (> 0) to consider when computing temporal difference (TD) targets. Controls the balance between TD learning and Monte Carlo methods: higher values reduce bias (by relying less on potentially inaccurate value estimates) but increase variance (by incorporating more environmental stochasticity and reducing the averaging effect). A value of 1 corresponds to standard TD learning with immediate bootstrapping, while very large values approach Monte Carlo-like estimation that uses complete episode returns.
target_update_freq – the number of training iterations between each complete update of the target network. Controls how frequently the target Q-network parameters are updated with the current Q-network values. A value of 0 disables the target network entirely, using only a single network for both action selection and bootstrap targets. Higher values provide more stable learning targets but slow down the propagation of new value estimates. Lower positive values allow faster learning but may lead to instability due to rapidly changing targets. Typically set between 100-10000 for DQN variants, with exact values depending on environment complexity.

fqf

Contents

fqf#