xenoverse 0.1.0__tar.gz

xenoverse-0.1.0/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   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.

xenoverse-0.1.0/PKG-INFO

@@ -0,0 +1,52 @@
+Metadata-Version: 2.1
+Name: xenoverse
+Version: 0.1.0
+Summary: Collection of xeno-world environments for meta-training of general-purpose learning agents (GLAs)
+Home-page: https://github.com/FutureAGI/xenoverse
+Author: WorldEditors
+Author-email: 
+License: Apache
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: gym>=0.18.0
+Requires-Dist: numpy>=1.24.4
+Requires-Dist: Pillow>=6.2.2
+Requires-Dist: six>=1.12.0
+Requires-Dist: pygame>=2.6.0
+Requires-Dist: numba>=0.58.1
+
+# L3C
+
+L3C provides variant environments facilitating In-Context Learning, featured in the following assets:
+- Vast amounts of diversified tasks with minimal inductive bias
+- Lifelong In-Context Learning
+- Generative and Interactive
+
+# Environments Updating
+
+- [AnyMDP](l3c/anymdp): Procedurally generated unlimited general-purpose Markov Decision Processes (MDP)
+
+- [MetaLanguage](l3c/metalang): Pseudo-language generated from randomized neural networks
+
+- [MazeWorld](l3c/mazeworld): Procedurally generated mazes with diverse maze structures, navigation goals to benchmark object navigation
+
+- [L3C_Baselines](https://github.com/FutureAGI/L3C_Baselines): An implementation of the baseline solutions to the above tasks
+
+# Installation
+
+```bash
+pip install l3c
+```
+
+# Reference
+Cite this work with
+
+```bibtex
+@article{wang2024benchmarking,
+  title={Benchmarking General Purpose In-Context Learning},
+  author={Wang, Fan and Lin, Chuan and Cao, Yang and Kang, Yu},
+  journal={arXiv preprint arXiv:2405.17234},
+  year={2024}
+}
+```

xenoverse-0.1.0/README.md

@@ -0,0 +1,34 @@
+# L3C
+
+L3C provides variant environments facilitating In-Context Learning, featured in the following assets:
+- Vast amounts of diversified tasks with minimal inductive bias
+- Lifelong In-Context Learning
+- Generative and Interactive
+
+# Environments Updating
+
+- [AnyMDP](l3c/anymdp): Procedurally generated unlimited general-purpose Markov Decision Processes (MDP)
+
+- [MetaLanguage](l3c/metalang): Pseudo-language generated from randomized neural networks
+
+- [MazeWorld](l3c/mazeworld): Procedurally generated mazes with diverse maze structures, navigation goals to benchmark object navigation
+
+- [L3C_Baselines](https://github.com/FutureAGI/L3C_Baselines): An implementation of the baseline solutions to the above tasks
+
+# Installation
+
+```bash
+pip install l3c
+```
+
+# Reference
+Cite this work with
+
+```bibtex
+@article{wang2024benchmarking,
+  title={Benchmarking General Purpose In-Context Learning},
+  author={Wang, Fan and Lin, Chuan and Cao, Yang and Kang, Yu},
+  journal={arXiv preprint arXiv:2405.17234},
+  year={2024}
+}
+```

xenoverse-0.1.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

xenoverse-0.1.0/setup.py

@@ -0,0 +1,51 @@
+#   Copyright (c) 2022 DeepEvolution 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.
+
+import io
+from xenoverse import __version__
+from setuptools import setup, find_packages
+
+with io.open('README.md', 'r', encoding='utf-8') as fh:
+    long_description = fh.read()
+
+setup(
+    name='xenoverse',
+    version=__version__,
+    author='WorldEditors',
+    author_email='',
+    description=('Collection of xeno-world environments for meta-training of general-purpose learning agents (GLAs)'),
+    long_description=long_description,
+    long_description_content_type='text/markdown',
+    url='https://github.com/FutureAGI/xenoverse',
+    license="Apache",
+    packages=[package for package in find_packages()
+              if package.startswith('xenoverse')],
+    package_data={'xenoverse': [
+        './mazeworld/envs/img/*',
+        ]
+    },
+    python_requires='>=3.8',
+    tests_require=['pytest', 'mock'],
+    include_package_data=True,
+    install_requires=[
+        'gym>=0.18.0',
+        'numpy>=1.24.4',
+        'Pillow>=6.2.2',
+        'six>=1.12.0',
+        'pygame>=2.6.0',
+        'numba>=0.58.1'
+    ],
+    extras_require={},
+    zip_safe=False,
+)

xenoverse-0.1.0/xenoverse/__init__.py

@@ -0,0 +1,15 @@
+#   Copyright (c) 2018 PaddlePaddle 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.
+
+__version__ = '0.1.0'

xenoverse-0.1.0/xenoverse/anymdp/__init__.py

@@ -0,0 +1,34 @@
+#   Copyright (c) 2018 PaddlePaddle 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.
+
+from gym.envs.registration import register
+from l3c.anymdp.anymdp_env import AnyMDPEnv, AnyMDPEnvD2C
+from l3c.anymdp.anymdp_solver_opt import AnyMDPSolverOpt
+from l3c.anymdp.anymdp_solver_ots import AnyMDPSolverOTS
+from l3c.anymdp.anymdp_solver_q import AnyMDPSolverQ
+from l3c.anymdp.task_sampler import AnyMDPTaskSampler
+from l3c.anymdp.visualizer import task_visualizer
+
+register(
+    id='anymdp-v0',
+    entry_point='l3c.anymdp:AnyMDPEnv',
+    kwargs={"max_steps": 5000},
+)
+
+register(
+    id='anymdp-d2c-v0',
+    entry_point='l3c.anymdp:AnyMDPEnvD2C',
+    kwargs={"max_steps": 5000,
+            "state_dim": 16},
+)

xenoverse-0.1.0/xenoverse/anymdp/anymdp_env.py

@@ -0,0 +1,123 @@
+"""
+Gym Environment For Any MDP
+"""
+import numpy
+import gym
+import pygame
+from numpy import random
+
+from gym import error, spaces, utils
+from gym.utils import seeding
+from l3c.utils import pseudo_random_seed
+
+class AnyMDPEnv(gym.Env):
+    def __init__(self, max_steps):
+        """
+        Pay Attention max_steps might be reseted by task settings
+        """
+        self.observation_space = spaces.Discrete(2)
+        self.action_space = spaces.Discrete(2)
+        self.max_steps = max_steps
+        self.task_set = False
+
+    def set_task(self, task_config):
+        self.transition_matrix = task_config["transition"]
+        self.reward_matrix = task_config["reward"]
+        self.reward_noise = task_config["reward_noise"]
+        self.reward_noise_type = task_config["reward_noise_type"]
+        self.state_mapping = task_config["state_mapping"]
+        self.reset_triggers = task_config["reset_triggers"]
+        self.reset_states = task_config["reset_states"]
+        self.n_states = task_config["state_space"]
+        self.n_actions = task_config["action_space"]
+
+        if("max_steps" in task_config):
+            self.max_steps = task_config["max_steps"]
+
+        ns1, na1, ns2 = self.transition_matrix.shape
+        ns3, na2, ns4 = self.reward_matrix.shape
+
+        assert ns1 == ns2 and ns2 == ns3 and ns3==ns4, \
+            "Transition matrix and reward matrix must have the same number of states"
+        assert na1 == na2, \
+            "Transition matrix and reward matrix must have the same number of actions"
+        assert ns1 > 0, "State space must be at least 1"
+        assert na1 > 1, "Action space must be at least 2"
+
+        self.observation_space = spaces.Discrete(self.n_states)
+        self.action_space = spaces.Discrete(self.n_actions)
+
+        self.task_set = True
+        self.need_reset = True
+
+    def reset(self):
+        if(not self.task_set):
+            raise Exception("Must call \"set_task\" first")
+        
+        self.steps = 0
+        self.need_reset = False
+        random.seed(pseudo_random_seed())
+
+        self._state = numpy.random.choice(len(self.reset_states),
+                                          replace=True,
+                                          p=self.reset_states)
+        return self.state_mapping[self._state], {"steps": self.steps}
+
+    def step(self, action):
+        if(self.need_reset or not self.task_set):
+            raise Exception("Must \"set_task\" and \"reset\" before doing any actions")
+        assert action < self.n_actions, "Action must be less than the number of actions"
+        transition_gt = self.transition_matrix[self._state, action]
+
+        next_state = random.choice(len(self.state_mapping), p=transition_gt)
+
+        reward_gt = self.reward_matrix[self._state, action, next_state]
+        reward_gt_noise = self.reward_noise[self._state, action, next_state]
+
+        if(self.reward_noise_type == 'normal'):
+            reward = random.normal(reward_gt, reward_gt_noise)
+        elif(self.reward_noise_type == 'binomial'):
+            if(reward_gt > 0):
+                reward = float(random.binomial(1, reward_gt))
+            else:
+                reward = - float(random.binomial(1, abs(reward_gt)))
+
+        info = {"steps": self.steps, "reward_gt": reward_gt}
+
+        if(self.state_mapping.ndim == 1):
+            transition_ext_gt = numpy.zeros((self.n_states,))
+            for i,s in enumerate(self.state_mapping):
+                transition_ext_gt[s] = transition_gt[i]
+            info["transition_gt"] = transition_ext_gt
+
+        self.steps += 1
+        self._state = next_state
+        #print("inner", next_state, "outer", self.state_mapping[next_state], self.max_steps,
+        #      "triggers", self.reset_triggers, "starts", self.reset_states)
+        done = (self.steps >= self.max_steps or self.reset_triggers[self._state])
+        if(done):
+            self.need_reset = True
+        return self.state_mapping[next_state], reward, done, info
+    
+    @property
+    def state(self):
+        return self.state_mapping[self._state]
+    
+    @property
+    def inner_state(self):
+        return self._state
+
+class AnyMDPEnvD2C(AnyMDPEnv):
+    """
+    Transfer a AnyMDPEnv to Continuous State Space without resampling a task
+    """
+    def __init__(self, max_steps, state_dim):
+        super().__init__(max_steps)
+        self.observation_space = spaces.Box(0., 1., shape=(state_dim,))
+        self.state_dim = state_dim
+
+    def set_task(self, task_config):
+        super().set_task(task_config)
+        n_inner_states = self.state_mapping.shape[0]
+        self.state_mapping = numpy.random.normal(size=(n_inner_states, self.state_dim))
+        self.observation_space = spaces.Box(0., 1., shape=(self.state_dim,))

xenoverse-0.1.0/xenoverse/anymdp/anymdp_solver_opt.py

@@ -0,0 +1,38 @@
+import gym
+import numpy
+from numpy import random
+from numba import njit
+from l3c.anymdp.solver import update_value_matrix, get_final_transition, get_final_reward
+
+
+class AnyMDPSolverOpt(object):
+    """
+    Solver for AnyMDPEnv with Bellman Equation and Value Iteration
+    Suppose to know the ground truth of the environment
+    """
+    def __init__(self, env, gamma=0.99):
+        if(not env.task_set):
+            raise Exception("AnyMDPEnv is not initialized by 'set_task', must call set_task first")
+        self.n_actions = env.action_space.n
+        self.n_states = env.observation_space.n
+        self.transition_matrix = get_final_transition(
+                transition=env.transition_matrix,
+                reset_states=env.reset_states,
+                reset_triggers=env.reset_triggers)
+        self.reward_matrix = get_final_reward(
+                reward=env.reward_matrix,
+                reset_triggers=env.reset_triggers,
+        )
+        self.state_mapping = env.state_mapping
+        self.value_matrix = numpy.zeros((self.n_states, self.n_actions))
+        self.gamma = gamma
+        self.inverse_state_mapping = dict()
+        for i,state in enumerate(self.state_mapping):
+            self.inverse_state_mapping[state] = i
+        self.q_solver()
+
+    def q_solver(self, gamma=0.99):
+        self.value_matrix = update_value_matrix(self.transition_matrix, self.reward_matrix, gamma, self.value_matrix)
+
+    def policy(self, state):
+        return numpy.argmax(self.value_matrix[self.inverse_state_mapping[state]])

xenoverse-0.1.0/xenoverse/anymdp/anymdp_solver_ots.py

@@ -0,0 +1,73 @@
+import gym
+import numpy
+from numpy import random
+
+
+class AnyMDPSolverOTS(object):
+    """
+    Implementing the RL Solver  of the paper
+    Hu, Bingshan, et al. "Optimistic Thompson sampling-based algorithms for 
+        episodic reinforcement learning." 
+        Uncertainty in Artificial Intelligence. PMLR, 2023.
+    """
+    def __init__(self, env, gamma=0.99, c=0.01, alpha=0.10, max_steps=4000):
+        """
+        The constructor for the class AnyMDPSolverQ
+        The exploration strategy is controlled by UCB-H with c as its hyperparameter. Increasing c will encourage exploration
+        Simulation of the ideal policy when the ground truth is not known
+        """
+        if(not env.task_set):
+            raise Exception("AnyMDPEnv is not initialized by 'set_task', must call set_task first")
+        self.n_actions = env.action_space.n
+        self.n_states = env.observation_space.n
+       
+        self.est_r = numpy.zeros((self.n_states, self.n_actions))
+        self.est_r_cnt = numpy.zeros((self.n_states, self.n_actions))
+        self.est_t = 0.1 * numpy.ones((self.n_states, self.n_actions, self.n_states))
+        self.gamma = gamma
+        self.lr = 1.0 - alpha
+        self._c = c / (1.0 - self.gamma)
+        self.max_steps = max_steps
+        self.reset_states = numpy.zeros(self.n_states)
+        for i,s in enumerate(env.state_mapping):
+            self.reset_states[s] = env.reset_states[i]
+
+        self.value_matrix = numpy.zeros((self.n_states, self.n_actions))
+        self.est_r_global_avg = 0
+        self.est_r_global_cnt = 0
+        self.alpha = alpha
+
+
+    def learner(self, s, a, ns, r, done):
+        # Update the environment model estimation
+        if(self.est_r_cnt[s,a] > 0):
+            self.est_r[s,a] += (r - self.est_r[s,a])/self.est_r_cnt[s,a]
+        else:
+            self.est_r[s,a] = r
+
+        if(self.est_r_global_cnt > 0):
+            self.est_r_global_avg += (r - self.est_r_global_avg) * self.alpha
+        else:
+            self.est_r_global_avg = r
+
+        self.est_r_cnt[s,a]+= 1
+        self.est_t[s,a,ns] += 1
+        self.est_r_global_cnt += 1
+
+        est_t = self.est_t / numpy.sum(self.est_t, axis=-1, keepdims=True)
+
+        if(not done):
+            est_target = numpy.sum(est_t[s, a] * numpy.max(self.value_matrix, axis=-1))
+            est_value = self.est_r[s,a] + self.gamma * est_target
+        else:
+            est_target = numpy.sum(self.reset_states * numpy.max(self.value_matrix, axis=-1))
+            est_value = self.est_r[s,a] + self.gamma * est_target
+        self.value_matrix[s,a] += self.alpha * (est_value - self.value_matrix[s,a])
+
+
+    def policy(self, state):
+        # Apply UCB with dynamic noise (Thompson Sampling)
+        values = self._c * numpy.sqrt(numpy.log(self.max_steps + 1) / numpy.clip(self.est_r_cnt[state], 1.0, None)) * \
+                numpy.maximum(numpy.random.randn(self.n_actions), 0) + \
+                self.value_matrix[state]
+        return numpy.argmax(values)

xenoverse-0.1.0/xenoverse/anymdp/anymdp_solver_q.py

@@ -0,0 +1,43 @@
+import gym
+import numpy
+from numpy import random
+
+class AnyMDPSolverQ(object):
+    """
+    Solver for AnyMDPEnv with Q-Learning
+    """
+    def __init__(self, env, gamma=0.99, c=0.01, alpha=0.01, max_steps=4000):
+        """
+        The constructor for the class AnyMDPSolverQ
+        The exploration strategy is controlled by UCB-H with c as its hyperparameter. Increasing c will encourage exploration
+        Simulation of the ideal policy when the ground truth is not known
+        """
+        if(not env.task_set):
+            raise Exception("AnyMDPEnv is not initialized by 'set_task', must call set_task first")
+        self.n_actions = env.action_space.n
+        self.n_states = env.observation_space.n
+        self.transition_matrix = env.transition_matrix
+        self.reward_matrix = env.reward_matrix
+        self.value_matrix = numpy.zeros((self.n_states, self.n_actions))
+        self.sa_vistied = numpy.zeros((self.n_states, self.n_actions))
+        self.gamma = gamma
+        self.alpha = alpha
+        self.max_steps = max_steps
+        self._c = c / (1.0 - self.gamma)
+
+
+    def learner(self, s, a, ns, r, done):
+        if(done):
+            target = r
+        else:
+            target = r + self.gamma * max(self.value_matrix[ns])
+        error = target - self.value_matrix[s][a]
+        self.value_matrix[s][a] += self.alpha * error
+        self.sa_vistied[s][a] += 1
+
+    def policy(self, state):
+        # Apply UCB with dynamic noise (Thompson Sampling)
+        values = self._c * numpy.sqrt(numpy.log(self.max_steps + 1) / numpy.clip(self.sa_vistied[state], 1.0, None)) * \
+                numpy.maximum(numpy.random.randn(self.n_actions), 0) + \
+                self.value_matrix[state]
+        return numpy.argmax(values)