├── README.md ├── figures ├── DQV-Max_estimates.png ├── DQV_estimates.png ├── dqv_max_boxing.jpg ├── dqv_max_enduro.jpg └── dqv_max_pong.jpg ├── models └── PongDeterministic-v4 │ ├── offline │ └── dqv-max │ │ ├── state_action_value_model.h5 │ │ └── state_value_model.h5 │ └── online │ └── dqv │ ├── state_action_value_model.h5 │ └── state_value_model.h5 ├── paper └── Arxiv_paper.pdf └── src ├── DQV_FULL_ATARI.sh ├── agent.py ├── choose_rl_ensemble.py ├── ddqn.py ├── dqn.py ├── dqv.py ├── dqv_max.py ├── duelling_dqv.py ├── test_value_functions.py ├── training_job.sh └── utils.py /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/README.md -------------------------------------------------------------------------------- /figures/DQV-Max_estimates.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/figures/DQV-Max_estimates.png -------------------------------------------------------------------------------- /figures/DQV_estimates.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/figures/DQV_estimates.png -------------------------------------------------------------------------------- /figures/dqv_max_boxing.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/figures/dqv_max_boxing.jpg -------------------------------------------------------------------------------- /figures/dqv_max_enduro.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/figures/dqv_max_enduro.jpg -------------------------------------------------------------------------------- /figures/dqv_max_pong.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/figures/dqv_max_pong.jpg -------------------------------------------------------------------------------- /models/PongDeterministic-v4/offline/dqv-max/state_action_value_model.h5: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/models/PongDeterministic-v4/offline/dqv-max/state_action_value_model.h5 -------------------------------------------------------------------------------- /models/PongDeterministic-v4/offline/dqv-max/state_value_model.h5: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/models/PongDeterministic-v4/offline/dqv-max/state_value_model.h5 -------------------------------------------------------------------------------- /models/PongDeterministic-v4/online/dqv/state_action_value_model.h5: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/models/PongDeterministic-v4/online/dqv/state_action_value_model.h5 -------------------------------------------------------------------------------- /models/PongDeterministic-v4/online/dqv/state_value_model.h5: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/models/PongDeterministic-v4/online/dqv/state_value_model.h5 -------------------------------------------------------------------------------- /paper/Arxiv_paper.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/paper/Arxiv_paper.pdf -------------------------------------------------------------------------------- /src/DQV_FULL_ATARI.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/src/DQV_FULL_ATARI.sh -------------------------------------------------------------------------------- /src/agent.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/src/agent.py -------------------------------------------------------------------------------- /src/choose_rl_ensemble.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/src/choose_rl_ensemble.py -------------------------------------------------------------------------------- /src/ddqn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/src/ddqn.py -------------------------------------------------------------------------------- /src/dqn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/src/dqn.py -------------------------------------------------------------------------------- /src/dqv.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/src/dqv.py -------------------------------------------------------------------------------- /src/dqv_max.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/src/dqv_max.py -------------------------------------------------------------------------------- /src/duelling_dqv.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/src/duelling_dqv.py -------------------------------------------------------------------------------- /src/test_value_functions.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/src/test_value_functions.py -------------------------------------------------------------------------------- /src/training_job.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/src/training_job.sh -------------------------------------------------------------------------------- /src/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/paintception/Deep-Quality-Value-Family/HEAD/src/utils.py --------------------------------------------------------------------------------