yardenas

import argparse
import timeit
from typing import Mapping, Union

import numpy as np

Transition = Mapping[str, Union[np.ndarray, dict]]
Batch = Mapping[str, np.ndarray]

yardenas

import argparse
import timeit
from typing import Mapping, Union

import jax
import jax.numpy as jnp
import numpy as np

Transition = Mapping[str, Union[np.ndarray, dict]]
Batch = Mapping[str, jnp.ndarray]

yardenas

"""
A bare bones examples of optimizing a black-box function (f) using
Natural Evolution Strategies (NES), where the parameter distribution is a 
gaussian of fixed standard deviation.
"""

import numpy as np
np.random.seed(0)

# the function we want to optimize

yardenas

""" Trains an agent with (stochastic) Policy Gradients on Pong. Uses OpenAI Gym. """
import numpy as np
import cPickle as pickle
import gym

# hyperparameters
H = 200 # number of hidden layer neurons
batch_size = 10 # every how many episodes to do a param update?
learning_rate = 1e-4
gamma = 0.99 # discount factor for reward