dennybritz/reinforcement-learning

Issues

• MC Control with Epsilon-Greedy Policies ---Epsilon Value and Best Action prob error

  #252 opened a year ago by hardik-kansal
  2

• Gambler's Problem: 0 Stake Allowed?

  #223 opened 5 years ago by mparigi
  1

• Questionable result in Gamblers Problem Solution

  #172 opened 6 years ago by bminixhofer
  12

• Reinforcement learning policy

  #238 opened 3 years ago by Comp-Engr18
  1

• demystifying-deep-reinforcement-learning link is broken

  #250 opened 2 years ago by kiankyars
  0

• Vanilla REINFORCE implementation

  #200 opened 6 years ago by alek5k
  2

• please provide requirements.txt or mention the exact version of packages used.

  #247 opened 2 years ago by Nahdus
  0

• Issue in: reinforcement-learning/MC/MC Prediction Solution.ipynb

  #246 opened 2 years ago by Almujtaba-Yaseen
  0

• Typo in: "Model-Free Prediction & Control with Monte Carlo (MC)" section -> "Blackjack Playground.ipynb" file:

  #244 opened 2 years ago by Almujtaba-Yaseen
  0

• A small correction in "MDPs and Bellman Equations" section

  #243 opened 2 years ago by Almujtaba-Yaseen
  0

• Batch update for Continuous Mountain Car Actor-Critic

  #180 opened 6 years ago by GoingMyWay
  1

• Policy Gradient Methods: Loss function of policy estimator in REINFORCE

  #181 opened 3 years ago by ArikVoronov
  3

• Minor Link fix

  #239 opened 3 years ago by gitDawn
  0

• Deep Q Learning, neither works with tensorflow 1.x nor with tensorflow 2.x

  #217 opened 5 years ago by azharsalman
  1

• DQN Testing Rewards on Atari Games

  #236 opened 4 years ago by willtop
  1

• Clarification on DQN testing rewards on Atari games

  #235 opened 4 years ago by willtop
  0

• Lecture Slides need an update

  #232 opened 4 years ago by harsh306
  0

• Randomness in optimal epsilon_greedy_policy

  #196 opened 4 years ago by levindabhi
  1

• Is the Implementation correct?

  #170 opened 4 years ago by Nerdyvedi
  0

• Monte Carlo AssertionError: defaultdict(<function mc_control_importance_sampling.<locals>.<lambda> at 0x7f31699ffe18>, {}) (<class 'collections.defaultdict'>)

  #231 opened 4 years ago by NC25
  0

• Policy Evaluation Exercise Solution Is Wrong

  #229 opened 4 years ago by ugrkm
  1

• DQL size error

  #227 opened 5 years ago by johan606303
  0

• Some question in MC Control with Epsilon-Greedy Policies Solution.ipynb

  #224 opened 5 years ago by josephbak
  2

• Why CliffWalkingEnv returns 'is_done=True' when reaching cliff?

  #219 opened 5 years ago by wakamori
  2

• Is a line missing in 'MC Control with Epsilon-Greedy Policies Solution.ipynb'?

  #220 opened 5 years ago by Ritz111
  1

• Policy iteration solution only show 1 optimal solution

  #212 opened 5 years ago by duongnhatthang
  2

• Provided policy_improvement() solution initializes values to zero for each iteration

  #204 opened 6 years ago by link2xt
  2

• why DQN use kernel size 8 ?

  #222 opened 5 years ago by opentld
  0

• Why is Chapter 11 excluded?

  #221 opened 5 years ago by BedirT
  2

• Can an agent learn valid actions offline, being able to choose only actions that were already taken (e.g. from historical data) ? [question]

  #218 opened 5 years ago by VieVaWaldi
  6

• Could anyone show me reason why use 4 same grayscale frames when training DQN?

  #213 opened 5 years ago by roachsinai
  1

• log

  #208 opened 5 years ago by Mahsa-Bastankhah
  0

• OSError: [Errno 12] Cannot allocate memory

  #173 opened 6 years ago by VictorLeeLk
  2

• Provided policy_improvement() solution is not guaranteed to terminate

  #203 opened 6 years ago by link2xt
  1

• policy_improvement() should be renamed to policy_iteration()

  #202 opened 6 years ago by link2xt
  0

• Blackjack - Monte Carlo Prediction

  #182 opened 6 years ago by rahulptel
  1

• Unstable reinforce with baseline model

  #192 opened 6 years ago by Jacobi93
  2

• feed action to critic network

  #190 opened 6 years ago by ehsaneshaghi
  0

• How to restore model

  #189 opened 6 years ago by tdr1991
  0

• The output layer should not using RELU activation function.

  #186 opened 6 years ago by wanjunhong0
  0

• Why RBFSampler from sklearn is used as the feature in the FA example?

  #184 opened 6 years ago by zyongxu
  0

• DQN Dense Tensor Using too Much Memory

  #183 opened 6 years ago by nflu
  1

• You don't follow the book?

  #185 opened 6 years ago by alexmosc
  0

• Define an envirement

  #179 opened 6 years ago by ewtrends
  0

• policy evaluation algorithm and implementation bug

  #177 opened 6 years ago by hamifthi
  6

• [bug] DQN/dqn.py: Incorrect loss function. [question] Question about RMSProp paramethers

  #174 opened 6 years ago by Kropekk
  5

• Policy Evaluation Solution VS Sutton's Page 75

  #171 opened 6 years ago by benjamintanweihao
  1

• Continuous MountainCar Actor Critic issue

  #168 opened 7 years ago by zhouPengF
  0

• Policy Gradient, when action space is 40, how can I sample action from Gaussian?

  #167 opened 7 years ago by GoingMyWay
  0

• eval() throws an error

  #161 opened 7 years ago by lechatthecat
  0