full_episode_objective.py

# Copyright 2017 The TensorFlow Authors All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================

"""Objectives for full-episode.

Implementations of UREX & REINFORCE.  Note that these implementations
use a non-parametric baseline to reduce variance.  Thus, multiple
samples with the same seed must be taken from the environment.

"""

import tensorflow as tf

import objective


class Reinforce(objective.Objective):
  def __init__(self, learning_rate, clip_norm, num_samples,
               tau=0.1, bonus_weight=1.0):
    super(Reinforce, self).__init__(learning_rate, clip_norm=clip_norm)
    self.num_samples = num_samples
    assert self.num_samples > 1
    self.tau = tau
    self.bonus_weight = bonus_weight
    self.eps_lambda = 0.0

  def get_bonus(self, total_rewards, total_log_probs):
    """Exploration bonus."""
    return -self.tau * total_log_probs

  def get(self, rewards, pads, values, final_values,
          log_probs, prev_log_probs, target_log_probs,
          entropies, logits,
          target_values, final_target_values):
    seq_length = tf.shape(rewards)[0]

    not_pad = tf.reshape(1 - pads, [seq_length, -1, self.num_samples])
    rewards = not_pad * tf.reshape(rewards, [seq_length, -1, self.num_samples])
    log_probs = not_pad * tf.reshape(sum(log_probs), [seq_length, -1, self.num_samples])

    total_rewards = tf.reduce_sum(rewards, 0)
    total_log_probs = tf.reduce_sum(log_probs, 0)

    rewards_and_bonus = (total_rewards +
                         self.bonus_weight *
                         self.get_bonus(total_rewards, total_log_probs))

    baseline = tf.reduce_mean(rewards_and_bonus, 1, keep_dims=True)

    loss = -tf.stop_gradient(rewards_and_bonus - baseline) * total_log_probs
    loss = tf.reduce_mean(loss)
    raw_loss = loss  # TODO

    gradient_ops = self.training_ops(
        loss, learning_rate=self.learning_rate)

    tf.summary.histogram('log_probs', total_log_probs)
    tf.summary.histogram('rewards', total_rewards)
    tf.summary.scalar('avg_rewards',
                      tf.reduce_mean(total_rewards))
    tf.summary.scalar('loss', loss)

    return loss, raw_loss, baseline, gradient_ops, tf.summary.merge_all()


class UREX(Reinforce):
  def get_bonus(self, total_rewards, total_log_probs):
    """Exploration bonus."""
    discrepancy = total_rewards / self.tau - total_log_probs
    normalized_d = self.num_samples * tf.nn.softmax(discrepancy)
    return self.tau * normalized_d