def setup_param_noise(self, normalized_obs0):
assert self.param_noise is not None
# Configure perturbed actor.
param_noise_actor = copy(self.actor)
param_noise_actor.name = 'param_noise_actor'
self.perturbed_actor_tf = param_noise_actor(normalized_obs0)
logger.info('setting up param noise')
self.perturb_policy_ops = get_perturbed_actor_updates(
self.actor, param_noise_actor, self.param_noise_stddev)
# Configure separate copy for stddev adoption.
adaptive_param_noise_actor = copy(self.actor)
adaptive_param_noise_actor.name = 'adaptive_param_noise_actor'
adaptive_actor_tf = adaptive_param_noise_actor(normalized_obs0)
self.perturb_adaptive_policy_ops = get_perturbed_actor_updates(
self.actor, adaptive_param_noise_actor, self.param_noise_stddev)
self.adaptive_policy_distance = tf.sqrt(
tf.reduce_mean(tf.square(self.actor_tf - adaptive_actor_tf)))
def getKfacPrecondUpdates(self, gradlist, varlist):
updatelist = []
vg = 0.
assert len(self.stats) > 0
assert len(self.stats_eigen) > 0
assert len(self.factors) > 0
counter = 0
grad_dict = {var: grad for grad, var in zip(gradlist, varlist)}
for grad, var in zip(gradlist, varlist):
GRAD_RESHAPE = False
GRAD_TRANSPOSE = False
fpropFactoredFishers = self.stats[var]['fprop_concat_stats']
bpropFactoredFishers = self.stats[var]['bprop_concat_stats']
if (len(fpropFactoredFishers) + len(bpropFactoredFishers)) > 0:
counter += 1
GRAD_SHAPE = grad.get_shape()
if len(grad.get_shape()) > 2:
# reshape conv kernel parameters
KW = int(grad.get_shape()[0])
KH = int(grad.get_shape()[1])
C = int(grad.get_shape()[2])
D = int(grad.get_shape()[3])
if len(fpropFactoredFishers) > 1 and self._channel_fac:
# reshape conv kernel parameters into tensor
grad = tf.reshape(grad, [KW * KH, C, D])
else:
# reshape conv kernel parameters into 2D grad
grad = tf.reshape(grad, [-1, D])
GRAD_RESHAPE = True
elif len(grad.get_shape()) == 1:
# reshape bias or 1D parameters
D = int(grad.get_shape()[0])
grad = tf.expand_dims(grad, 0)
GRAD_RESHAPE = True
else:
# 2D parameters
C = int(grad.get_shape()[0])
D = int(grad.get_shape()[1])
if (self.stats[var]['assnBias']
is not None) and not self._blockdiag_bias:
# use homogeneous coordinates only works for 2D grad.
# TO-DO: figure out how to factorize bias grad
# stack bias grad
var_assnBias = self.stats[var]['assnBias']
grad = tf.concat(
[grad,
tf.expand_dims(grad_dict[var_assnBias], 0)], 0)
# project gradient to eigen space and reshape the eigenvalues
# for broadcasting
eigVals = []
for idx, stats in enumerate(
self.stats[var]['fprop_concat_stats']):
Q = self.stats_eigen[stats]['Q']
e = detectMinVal(self.stats_eigen[stats]['e'],
var,
name='act',
debug=KFAC_DEBUG)
Q, e = factorReshape(Q, e, grad, facIndx=idx, ftype='act')
eigVals.append(e)
grad = gmatmul(Q, grad, transpose_a=True, reduce_dim=idx)
for idx, stats in enumerate(
self.stats[var]['bprop_concat_stats']):
Q = self.stats_eigen[stats]['Q']
e = detectMinVal(self.stats_eigen[stats]['e'],
var,
name='grad',
debug=KFAC_DEBUG)
Q, e = factorReshape(Q, e, grad, facIndx=idx, ftype='grad')
eigVals.append(e)
grad = gmatmul(grad, Q, transpose_b=False, reduce_dim=idx)
##
#####
# whiten using eigenvalues
weightDecayCoeff = 0.
if var in self._weight_decay_dict:
weightDecayCoeff = self._weight_decay_dict[var]
if KFAC_DEBUG:
print(('weight decay coeff for %s is %f' %
(var.name, weightDecayCoeff)))
if self._factored_damping:
if KFAC_DEBUG:
print(('use factored damping for %s' % (var.name)))
coeffs = 1.
num_factors = len(eigVals)
# compute the ratio of two trace norm of the left and right
# KFac matrices, and their generalization
if len(eigVals) == 1:
damping = self._epsilon + weightDecayCoeff
else:
damping = tf.pow(self._epsilon + weightDecayCoeff,
1. / num_factors)
eigVals_tnorm_avg = [
tf.reduce_mean(tf.abs(e)) for e in eigVals
]
for e, e_tnorm in zip(eigVals, eigVals_tnorm_avg):
eig_tnorm_negList = [
item for item in eigVals_tnorm_avg
if item != e_tnorm
]
if len(eigVals) == 1:
adjustment = 1.
elif len(eigVals) == 2:
adjustment = tf.sqrt(e_tnorm /
eig_tnorm_negList[0])
else:
eig_tnorm_negList_prod = reduce(
lambda x, y: x * y, eig_tnorm_negList)
adjustment = tf.pow(
tf.pow(e_tnorm, num_factors - 1.) /
eig_tnorm_negList_prod, 1. / num_factors)
coeffs *= (e + adjustment * damping)
else:
coeffs = 1.
damping = (self._epsilon + weightDecayCoeff)
for e in eigVals:
coeffs *= e
coeffs += damping
#grad = tf.Print(grad, [tf.convert_to_tensor('1'), tf.convert_to_tensor(var.name), grad.get_shape()])
grad /= coeffs
#grad = tf.Print(grad, [tf.convert_to_tensor('2'), tf.convert_to_tensor(var.name), grad.get_shape()])
#####
# project gradient back to euclidean space
for idx, stats in enumerate(
self.stats[var]['fprop_concat_stats']):
Q = self.stats_eigen[stats]['Q']
grad = gmatmul(Q, grad, transpose_a=False, reduce_dim=idx)
for idx, stats in enumerate(
self.stats[var]['bprop_concat_stats']):
Q = self.stats_eigen[stats]['Q']
grad = gmatmul(grad, Q, transpose_b=True, reduce_dim=idx)
##
#grad = tf.Print(grad, [tf.convert_to_tensor('3'), tf.convert_to_tensor(var.name), grad.get_shape()])
if (self.stats[var]['assnBias']
is not None) and not self._blockdiag_bias:
# use homogeneous coordinates only works for 2D grad.
# TO-DO: figure out how to factorize bias grad
# un-stack bias grad
var_assnBias = self.stats[var]['assnBias']
C_plus_one = int(grad.get_shape()[0])
grad_assnBias = tf.reshape(
tf.slice(grad, begin=[C_plus_one - 1, 0], size=[1,
-1]),
var_assnBias.get_shape())
grad_assnWeights = tf.slice(grad,
begin=[0, 0],
size=[C_plus_one - 1, -1])
grad_dict[var_assnBias] = grad_assnBias
grad = grad_assnWeights
#grad = tf.Print(grad, [tf.convert_to_tensor('4'), tf.convert_to_tensor(var.name), grad.get_shape()])
if GRAD_RESHAPE:
grad = tf.reshape(grad, GRAD_SHAPE)
grad_dict[var] = grad
print(('projecting %d gradient matrices' % counter))
for g, var in zip(gradlist, varlist):
grad = grad_dict[var]
### clipping ###
if KFAC_DEBUG:
print(('apply clipping to %s' % (var.name)))
tf.Print(grad, [tf.sqrt(tf.reduce_sum(tf.pow(grad, 2)))],
"Euclidean norm of new grad")
local_vg = tf.reduce_sum(grad * g * (self._lr * self._lr))
vg += local_vg
# recale everything
if KFAC_DEBUG:
print('apply vFv clipping')
scaling = tf.minimum(1., tf.sqrt(self._clip_kl / vg))
if KFAC_DEBUG:
scaling = tf.Print(scaling, [
tf.convert_to_tensor('clip: '), scaling,
tf.convert_to_tensor(' vFv: '), vg
])
with tf.control_dependencies([tf.assign(self.vFv, vg)]):
updatelist = [grad_dict[var] for var in varlist]
for i, item in enumerate(updatelist):
updatelist[i] = scaling * item
return updatelist
def compute_stats(self, loss_sampled, var_list=None):
varlist = var_list
if varlist is None:
varlist = tf.trainable_variables()
gs = tf.gradients(loss_sampled, varlist, name='gradientsSampled')
self.gs = gs
factors = self.getFactors(gs, varlist)
stats = self.getStats(factors, varlist)
updateOps = []
statsUpdates = {}
statsUpdates_cache = {}
for var in varlist:
opType = factors[var]['opName']
fops = factors[var]['op']
fpropFactor = factors[var]['fpropFactors_concat']
fpropStats_vars = stats[var]['fprop_concat_stats']
bpropFactor = factors[var]['bpropFactors_concat']
bpropStats_vars = stats[var]['bprop_concat_stats']
SVD_factors = {}
for stats_var in fpropStats_vars:
stats_var_dim = int(stats_var.get_shape()[0])
if stats_var not in statsUpdates_cache:
old_fpropFactor = fpropFactor
B = (tf.shape(fpropFactor)[0]) # batch size
if opType == 'Conv2D':
strides = fops.get_attr("strides")
padding = fops.get_attr("padding")
convkernel_size = var.get_shape()[0:3]
KH = int(convkernel_size[0])
KW = int(convkernel_size[1])
C = int(convkernel_size[2])
flatten_size = int(KH * KW * C)
Oh = int(bpropFactor.get_shape()[1])
Ow = int(bpropFactor.get_shape()[2])
if Oh == 1 and Ow == 1 and self._channel_fac:
# factorization along the channels
# assume independence among input channels
# factor = B x 1 x 1 x (KH xKW x C)
# patches = B x Oh x Ow x (KH xKW x C)
if len(SVD_factors) == 0:
if KFAC_DEBUG:
print((
'approx %s act factor with rank-1 SVD factors'
% (var.name)))
# find closest rank-1 approx to the feature map
S, U, V = tf.batch_svd(
tf.reshape(fpropFactor, [-1, KH * KW, C]))
# get rank-1 approx slides
sqrtS1 = tf.expand_dims(tf.sqrt(S[:, 0, 0]), 1)
patches_k = U[:, :, 0] * sqrtS1 # B x KH*KW
full_factor_shape = fpropFactor.get_shape()
patches_k.set_shape(
[full_factor_shape[0], KH * KW])
patches_c = V[:, :, 0] * sqrtS1 # B x C
patches_c.set_shape([full_factor_shape[0], C])
SVD_factors[C] = patches_c
SVD_factors[KH * KW] = patches_k
fpropFactor = SVD_factors[stats_var_dim]
else:
# poor mem usage implementation
patches = tf.extract_image_patches(
fpropFactor,
ksizes=[
1, convkernel_size[0], convkernel_size[1],
1
],
strides=strides,
rates=[1, 1, 1, 1],
padding=padding)
if self._approxT2:
if KFAC_DEBUG:
print(('approxT2 act fisher for %s' %
(var.name)))
# T^2 terms * 1/T^2, size: B x C
fpropFactor = tf.reduce_mean(patches, [1, 2])
else:
# size: (B x Oh x Ow) x C
fpropFactor = tf.reshape(
patches, [-1, flatten_size]) / Oh / Ow
fpropFactor_size = int(fpropFactor.get_shape()[-1])
if stats_var_dim == (fpropFactor_size +
1) and not self._blockdiag_bias:
if opType == 'Conv2D' and not self._approxT2:
# correct padding for numerical stability (we
# divided out OhxOw from activations for T1 approx)
fpropFactor = tf.concat([
fpropFactor,
tf.ones([tf.shape(fpropFactor)[0], 1]) / Oh /
Ow
], 1)
else:
# use homogeneous coordinates
fpropFactor = tf.concat([
fpropFactor,
tf.ones([tf.shape(fpropFactor)[0], 1])
], 1)
# average over the number of data points in a batch
# divided by B
cov = tf.matmul(fpropFactor, fpropFactor,
transpose_a=True) / tf.cast(B, tf.float32)
updateOps.append(cov)
statsUpdates[stats_var] = cov
if opType != 'Conv2D':
# HACK: for convolution we recompute fprop stats for
# every layer including forking layers
statsUpdates_cache[stats_var] = cov
for stats_var in bpropStats_vars:
stats_var_dim = int(stats_var.get_shape()[0])
if stats_var not in statsUpdates_cache:
old_bpropFactor = bpropFactor
bpropFactor_shape = bpropFactor.get_shape()
B = tf.shape(bpropFactor)[0] # batch size
C = int(bpropFactor_shape[-1]) # num channels
if opType == 'Conv2D' or len(bpropFactor_shape) == 4:
if fpropFactor is not None:
if self._approxT2:
if KFAC_DEBUG:
print(('approxT2 grad fisher for %s' %
(var.name)))
bpropFactor = tf.reduce_sum(
bpropFactor, [1, 2]) # T^2 terms * 1/T^2
else:
bpropFactor = tf.reshape(
bpropFactor,
[-1, C]) * Oh * Ow # T * 1/T terms
else:
# just doing block diag approx. spatial independent
# structure does not apply here. summing over
# spatial locations
if KFAC_DEBUG:
print(('block diag approx fisher for %s' %
(var.name)))
bpropFactor = tf.reduce_sum(bpropFactor, [1, 2])
# assume sampled loss is averaged. TO-DO:figure out better
# way to handle this
bpropFactor *= tf.to_float(B)
##
cov_b = tf.matmul(bpropFactor,
bpropFactor,
transpose_a=True) / tf.to_float(
tf.shape(bpropFactor)[0])
updateOps.append(cov_b)
statsUpdates[stats_var] = cov_b
statsUpdates_cache[stats_var] = cov_b
if KFAC_DEBUG:
aKey = list(statsUpdates.keys())[0]
statsUpdates[aKey] = tf.Print(statsUpdates[aKey], [
tf.convert_to_tensor('step:'),
self.global_step,
tf.convert_to_tensor('computing stats'),
])
self.statsUpdates = statsUpdates
return statsUpdates
def reduce_std(x, axis=None, keep_dims=False):
return tf.sqrt(reduce_var(x, axis=axis, keep_dims=keep_dims))
def __init__(self, size, eps=1e-2, default_clip_range=np.inf, sess=None):
"""A normalizer that ensures that observations are approximately distributed according to
a standard Normal distribution (i.e. have mean zero and variance one).
Args:
size (int): the size of the observation to be normalized
eps (float): a small constant that avoids underflows
default_clip_range (float): normalized observations are clipped to be in
[-default_clip_range, default_clip_range]
sess (object): the TensorFlow session to be used
"""
self.size = size
self.eps = eps
self.default_clip_range = default_clip_range
self.sess = sess if sess is not None else tf.get_default_session()
self.local_sum = np.zeros(self.size, np.float32)
self.local_sumsq = np.zeros(self.size, np.float32)
self.local_count = np.zeros(1, np.float32)
self.sum_tf = tf.get_variable(initializer=tf.zeros_initializer(),
shape=self.local_sum.shape,
name='sum',
trainable=False,
dtype=tf.float32)
self.sumsq_tf = tf.get_variable(initializer=tf.zeros_initializer(),
shape=self.local_sumsq.shape,
name='sumsq',
trainable=False,
dtype=tf.float32)
self.count_tf = tf.get_variable(initializer=tf.ones_initializer(),
shape=self.local_count.shape,
name='count',
trainable=False,
dtype=tf.float32)
self.mean = tf.get_variable(initializer=tf.zeros_initializer(),
shape=(self.size, ),
name='mean',
trainable=False,
dtype=tf.float32)
self.std = tf.get_variable(initializer=tf.ones_initializer(),
shape=(self.size, ),
name='std',
trainable=False,
dtype=tf.float32)
self.count_pl = tf.placeholder(name='count_pl',
shape=(1, ),
dtype=tf.float32)
self.sum_pl = tf.placeholder(name='sum_pl',
shape=(self.size, ),
dtype=tf.float32)
self.sumsq_pl = tf.placeholder(name='sumsq_pl',
shape=(self.size, ),
dtype=tf.float32)
self.update_op = tf.group(self.count_tf.assign_add(self.count_pl),
self.sum_tf.assign_add(self.sum_pl),
self.sumsq_tf.assign_add(self.sumsq_pl))
self.recompute_op = tf.group(
tf.assign(self.mean, self.sum_tf / self.count_tf),
tf.assign(
self.std,
tf.sqrt(
tf.maximum(
tf.square(self.eps), self.sumsq_tf / self.count_tf -
tf.square(self.sum_tf / self.count_tf)))),
)
self.lock = threading.Lock()