Denis Yarats denisyarats

PhD student in AI at New York University

denisyarats / on_policy_mc.py

Created January 15, 2017 05:32

on policy mc

	#!/usr/local/bin/python

	import argparse
	import numpy as np
	from collections import defaultdict
	import gym
	from gym import wrappers
	import pdb

	EXP_NAME_PREFIX = 'exp/on_policy_mc'

denisyarats / gist:4981579b42c8a08e49206347f7d28c6c

Created January 17, 2017 07:16

./sarsa.py --max_episodes 10000 --alpha 0.3 --gamma 0.9 --eps 0.2 --eps_schedule 200 --goal 25 --env copy --upload

denisyarats / q_learnig.py

Created January 19, 2017 06:46

	#!/usr/local/bin/python
	"""
	Q-learning - off policy TD(0) learning.

	Q(S, A) <- Q(S, A) + alpha * ((R + gamma * max(Q(S', A'))) - Q(S, A))
	A ~ e-greedy from pi(A\|S)
	"""

	import argparse
	import numpy as np

denisyarats / q_learnig.py

Created January 19, 2017 07:55

	#!/usr/local/bin/python
	"""
	Q-learning - off policy TD(0) learning.

	Q(S, A) <- Q(S, A) + alpha * ((R + gamma * max(Q(S', A'))) - Q(S, A))
	A ~ e-greedy from pi(A\|S)
	"""

	import argparse
	import numpy as np

denisyarats / q_learning_lin_appr.py

Created January 24, 2017 07:06

q-learning with linear approximation

	#!/usr/local/bin/python
	"""
	Q-learning with value fucntion approximation
	"""

	import argparse
	import numpy as np
	import matplotlib
	from matplotlib import pyplot as plt
	from mpl_toolkits.mplot3d import Axes3D

denisyarats / reinforce.py

Last active March 22, 2017 15:11

	import argparse
	import pdb

	import torch
	import torch.nn as nn
	import torch.optim as optim
	from torch.autograd import Variable
	import torch.nn.functional as F

	import numpy as np

denisyarats / pytorch_dqn.py

Created April 6, 2017 06:59

	#!/usr/bin/env python
	"""
	PyTorch implementation of DQN
	Paper: https://www.cs.toronto.edu/~vmnih/docs/dqn.pdf

	"""

	import argparse
	import gym
	from gym import wrappers

denisyarats / layernorm_gru.py

Created July 7, 2017 00:14

LayerNorm GRU

	import torch.nn as nn
	import torch.nn.functional as F
	import torch.nn.init
	from torch.autograd import Variable

	from models.utils import *


	class LayerNormGRUCell(nn.GRUCell):
	def __init__(self, input_size, hidden_size, bias=True):

denisyarats / py

Created July 3, 2019 15:21

seed

denisyarats / seed.py

Created July 3, 2019 15:21