def default_config() -> ClientConfig:
"""
:return: Default configuration for the experiment
"""
dqn_config = DQNConfig(input_dim=242,
defender_output_dim=242,
hidden_dim=64,
replay_memory_size=10000,
num_hidden_layers=1,
replay_start_size=1000,
batch_size=32,
target_network_update_freq=1000,
gpu=True,
tensorboard=True,
tensorboard_dir=default_output_dir() +
"/results/tensorboard",
loss_fn="Huber",
optimizer="Adam",
lr_exp_decay=True,
lr_decay_rate=0.9999)
q_agent_config = QAgentConfig(
gamma=0.999,
alpha=0.00001,
epsilon=1,
render=False,
eval_sleep=0.9,
min_epsilon=0.01,
eval_episodes=100,
train_log_frequency=100,
epsilon_decay=0.9999,
video=True,
eval_log_frequency=1,
video_fps=5,
video_dir=default_output_dir() + "/results/videos",
num_episodes=20001,
eval_render=False,
gifs=True,
gif_dir=default_output_dir() + "/results/gifs",
eval_frequency=1000,
attacker=False,
defender=True,
video_frequency=101,
save_dir=default_output_dir() + "/results/data",
dqn_config=dqn_config,
checkpoint_freq=5000)
env_name = "idsgame-random_attack-v4"
client_config = ClientConfig(env_name=env_name,
defender_type=AgentType.DQN_AGENT.value,
mode=RunnerMode.TRAIN_DEFENDER.value,
q_agent_config=q_agent_config,
output_dir=default_output_dir(),
title="RandomAttacker vs DQN",
run_many=True,
random_seeds=[0, 999, 299, 399, 499])
return client_config
def default_config() -> ClientConfig:
"""
:return: Default configuration for the experiment
"""
dqn_config = DQNConfig(input_dim=242,
attacker_output_dim=220,
hidden_dim=64,
replay_memory_size=10000,
num_hidden_layers=1,
replay_start_size=1000,
batch_size=32,
target_network_update_freq=5000,
gpu=True,
tensorboard=True,
tensorboard_dir=default_output_dir() +
"/tensorboard",
loss_fn="Huber",
optimizer="Adam",
lr_exp_decay=True,
lr_decay_rate=0.99995)
q_agent_config = QAgentConfig(gamma=1,
alpha=0.00001,
epsilon=0.75,
render=False,
eval_sleep=0.9,
min_epsilon=0.05,
eval_episodes=100,
train_log_frequency=100,
epsilon_decay=0.99995,
video=True,
eval_log_frequency=1,
video_fps=5,
video_dir=default_output_dir() + "/videos",
num_episodes=500000,
eval_render=False,
gifs=True,
gif_dir=default_output_dir() + "/gifs",
eval_frequency=5000,
attacker=True,
defender=False,
video_frequency=101,
save_dir=default_output_dir() + "/data",
dqn_config=dqn_config,
checkpoint_freq=10000)
env_name = "idsgame-minimal_defense-v6"
client_config = ClientConfig(
env_name=env_name,
attacker_type=AgentType.DQN_AGENT.value,
mode=RunnerMode.TRAIN_ATTACKER.value,
q_agent_config=q_agent_config,
output_dir=default_output_dir(),
title="TrainingDQNAgent vs DefendMinimalDefender")
return client_config
def test_train_random_attack_tabular_q_learning(version) -> ClientConfig:
q_agent_config = QAgentConfig(num_episodes=10,
eval_frequency=100,
attacker=False,
defender=True)
env_name = "idsgame-random_attack-v" + str(version)
client_config = ClientConfig(env_name=env_name,
defender_type=AgentType.TABULAR_Q_AGENT.value,
mode=RunnerMode.TRAIN_DEFENDER.value,
q_agent_config=q_agent_config,
output_dir=default_output_dir())
Runner.run(client_config)
def default_config() -> ClientConfig:
"""
:return: Default configuration for the experiment
"""
simulation_config = SimulationConfig(render=True, sleep=0.8, video=True, log_frequency=1,
video_fps=5, video_dir=default_output_dir() + "/videos", num_episodes=1000,
gifs=True, gif_dir=default_output_dir() + "/gifs", video_frequency = 1)
q_agent_config = QAgentConfig(attacker_load_path=default_output_dir() + "/q_table/q_table.npy")
env_name = "idsgame-v4"
client_config = ClientConfig(env_name=env_name, attacker_type=AgentType.TABULAR_Q_AGENT.value,
defender_type=AgentType.DEFEND_MINIMAL_VALUE.value, mode=RunnerMode.SIMULATE.value,
simulation_config=simulation_config, output_dir=default_output_dir(),
title="TabularQAgentAttacker vs DefendMinimalDefender",
q_agent_config=q_agent_config,
initial_state_path = default_output_dir() + "/initial_state/initial_state.pkl")
return client_config
def default_config() -> ClientConfig:
"""
:return: Default configuration for the experiment
"""
q_agent_config = QAgentConfig(gamma=0.999, alpha=0.05, epsilon=1, render=False, eval_sleep=0.9,
min_epsilon=0.01, eval_episodes=100, train_log_frequency=1,
epsilon_decay=0.9999, video=True, eval_log_frequency=1,
video_fps=5, video_dir=default_output_dir() + "/videos", num_episodes=40000,
eval_render=False, gifs=True, gif_dir=default_output_dir() + "/gifs",
eval_frequency=5000, attacker=True, defender=False, video_frequency=101,
save_dir=default_output_dir() + "/data")
env_name = "idsgame-minimal_defense-v4"
client_config = ClientConfig(env_name=env_name, attacker_type=AgentType.TABULAR_Q_AGENT.value,
mode=RunnerMode.TRAIN_ATTACKER.value,
q_agent_config=q_agent_config, output_dir=default_output_dir(),
title="TrainingQAgent vs DefendMinimalDefender")
return client_config
def default_config() -> ClientConfig:
"""
:return: Default configuration for the experiment
"""
q_agent_config = QAgentConfig(gamma=0.999, alpha=0.0005, epsilon=1, render=False, eval_sleep=0.9,
min_epsilon=0.01, eval_episodes=100, train_log_frequency=100,
epsilon_decay=0.9999, video=True, eval_log_frequency=1,
video_fps=5, video_dir=default_output_dir() + "/videos", num_episodes=20001,
eval_render=False, gifs=True, gif_dir=default_output_dir() + "/gifs",
eval_frequency=1000, attacker=False, defender=True,
save_dir=default_output_dir() + "/data", video_frequency=101)
env_name = "idsgame-maximal_attack-v7"
client_config = ClientConfig(env_name=env_name, defender_type=AgentType.TABULAR_Q_AGENT.value,
mode=RunnerMode.TRAIN_DEFENDER.value,
q_agent_config=q_agent_config, output_dir=default_output_dir(),
title="AttackMaximalAttacker vs TrainingQAgent",
run_many=True, random_seeds=[0, 999, 299, 399, 499])
return client_config
def default_config() -> ClientConfig:
"""
:return: Default configuration for the experiment
"""
q_agent_config = QAgentConfig(
gamma=0.999,
alpha=0.0005,
epsilon=1,
render=False,
eval_sleep=0.9,
min_epsilon=0.01,
eval_episodes=100,
train_log_frequency=100,
epsilon_decay=0.999,
video=True,
eval_log_frequency=1,
video_fps=5,
video_dir=default_output_dir() + "/results/videos",
num_episodes=20001,
eval_render=False,
gifs=True,
gif_dir=default_output_dir() + "/results/gifs",
eval_frequency=2000,
attacker=True,
defender=False,
video_frequency=101,
save_dir=default_output_dir() + "/results/data",
tab_full_state_space=True)
env_name = "idsgame-minimal_defense-v13"
client_config = ClientConfig(
env_name=env_name,
attacker_type=AgentType.TABULAR_Q_AGENT.value,
mode=RunnerMode.TRAIN_ATTACKER.value,
q_agent_config=q_agent_config,
output_dir=default_output_dir(),
title="TrainingQAgent vs DefendMinimalDefender",
run_many=False,
random_seeds=[0, 999, 299, 399, 499])
return client_config
optimizer="Adam",
lr_exp_decay=True,
lr_decay_rate=0.9999)
q_agent_config = QAgentConfig(
gamma=0.999,
alpha=0.00001,
epsilon=1,
render=False,
eval_sleep=0.9,
min_epsilon=0.01,
eval_episodes=100,
train_log_frequency=100,
epsilon_decay=0.9999,
video=True,
eval_log_frequency=1,
video_fps=5,
video_dir=default_output_dir() + "/results/videos/" + str(random_seed),
num_episodes=20001,
eval_render=False,
gifs=True,
gif_dir=default_output_dir() + "/results/gifs/" + str(random_seed),
eval_frequency=1000,
attacker=False,
defender=True,
video_frequency=101,
save_dir=default_output_dir() + "/results/data/" + str(random_seed),
dqn_config=dqn_config,
checkpoint_freq=5000)
env_name = "idsgame-maximal_attack-v3"
env = gym.make(env_name,
save_dir=default_output_dir() + "/results/data/" +
