-
Notifications
You must be signed in to change notification settings - Fork 11
/
min_norm_solvers.py
206 lines (179 loc) · 7.61 KB
/
min_norm_solvers.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
import numpy as np
import torch
class MinNormSolver:
MAX_ITER = 250
STOP_CRIT = 1e-5
def _min_norm_element_from2(v1v1, v1v2, v2v2):
'''
Analytical solution for min_{c} |cx_1 + (1-c)x_2|_2^2
d is the distance (objective) optimzed
v1v1 = <x1,x1>
v1v2 = <x1,x2>
v2v2 = <x2,x2>
'''
if v1v2 >= v1v1:
# Case: Fig 1, third column
gamma = 0.999
cost = v1v1
return gamma, cost
if v1v2 >= v2v2:
# Case: Fig 1, first column
gamma = 0.001
cost = v2v2
return gamma, cost
# Case: Fig 1, second column
gamma = -1.0 * ( (v1v2 - v2v2) / (v1v1+v2v2 - 2*v1v2) )
cost = v2v2 + gamma*(v1v2 - v2v2)
return gamma, cost
def _min_norm_2d(vecs, dps):
'''
Find the minimum norm solution as combination of two points
This is correct only in 2D
ie. min_c |\sum c_i x_i|_2^2 st. \sum c_i = 1 , 1 >= c_1 >= 0 for all i, c_i + c_j = 1.0 for some i, j
'''
dmin = 1e8
for i in range(len(vecs)):
for j in range(i+1,len(vecs)):
if (i,j) not in dps:
dps[(i, j)] = 0.0
for k in range(len(vecs[i])):
dps[(i,j)] += torch.mul(vecs[i][k], vecs[j][k]).sum().data.cpu()
dps[(j, i)] = dps[(i, j)]
if (i,i) not in dps:
dps[(i, i)] = 0.0
for k in range(len(vecs[i])):
dps[(i,i)] += torch.mul(vecs[i][k], vecs[i][k]).sum().data.cpu()
if (j,j) not in dps:
dps[(j, j)] = 0.0
for k in range(len(vecs[i])):
dps[(j, j)] += torch.mul(vecs[j][k], vecs[j][k]).sum().data.cpu()
c,d = MinNormSolver._min_norm_element_from2(dps[(i,i)], dps[(i,j)], dps[(j,j)])
if d < dmin:
dmin = d
sol = [(i,j),c,d]
return sol, dps
def _projection2simplex(y):
'''
Given y, it solves argmin_z |y-z|_2 st \sum z = 1 , 1 >= z_i >= 0 for all i
'''
m = len(y)
sorted_y = torch.from_numpy(np.flip(np.sort(y), axis=0).copy())
tmpsum = 0.0
tmax_f = (torch.sum(y) - 1.0)/m
for i in range(m-1):
tmpsum+= sorted_y[i]
tmax = (tmpsum - 1)/ (i+1.0)
if tmax > sorted_y[i+1]:
tmax_f = tmax
break
return torch.maximum(y - tmax_f, torch.zeros(y.shape))
def _next_point(cur_val, grad, n):
if type(cur_val) is np.ndarray:
cur_val = torch.Tensor(cur_val)
grad = torch.Tensor(grad)
proj_grad = grad - ( torch.sum(grad) / n )
tm1 = -1.0 * cur_val[proj_grad<0]/proj_grad[proj_grad<0]
tm2 = 1.0 - cur_val[proj_grad>0]/proj_grad[proj_grad>0]
# if type(tm1) is torch.Tensor:
# tm1 = tm1.numpy()
# tm2 = tm2.numpy()
skippers = torch.sum(tm1<1e-7) + torch.sum(tm2<1e-7)
t = 1
if len(tm1[tm1>1e-7]) > 0:
t = torch.min(tm1[tm1>1e-7])
if len(tm2[tm2>1e-7]) > 0:
t = min(t, torch.min(tm2[tm2>1e-7]))
next_point = proj_grad*t + cur_val
next_point = MinNormSolver._projection2simplex(next_point)
return next_point
def find_min_norm_element(vecs):
'''
Given a list of vectors (vecs), this method finds the minimum norm element in the convex hull
as min |u|_2 st. u = \sum c_i vecs[i] and \sum c_i = 1.
It is quite geometric, and the main idea is the fact that if d_{ij} = min |u|_2 st u = c x_i + (1-c) x_j; the solution lies in (0, d_{i,j})
Hence, we find the best 2-task solution, and then run the projected gradient descent until convergence
'''
# Solution lying at the combination of two points
dps = {}
init_sol, dps = MinNormSolver._min_norm_2d(vecs, dps)
n=len(vecs)
sol_vec = torch.zeros(n)
sol_vec[init_sol[0][0]] = init_sol[1]
sol_vec[init_sol[0][1]] = 1 - init_sol[1]
if n < 3:
# This is optimal for n=2, so return the solution
return sol_vec , init_sol[2]
iter_count = 0
grad_mat = torch.zeros((n,n))
for i in range(n):
for j in range(n):
grad_mat[i,j] = dps[(i, j)]
while iter_count < MinNormSolver.MAX_ITER:
grad_dir = -1.0* torch.from_numpy(np.dot(grad_mat, sol_vec))
new_point = MinNormSolver._next_point(sol_vec, grad_dir, n)
# Re-compute the inner products for line search
v1v1 = 0.0
v1v2 = 0.0
v2v2 = 0.0
for i in range(n):
for j in range(n):
v1v1 += sol_vec[i]*sol_vec[j]*dps[(i,j)]
v1v2 += sol_vec[i]*new_point[j]*dps[(i,j)]
v2v2 += new_point[i]*new_point[j]*dps[(i,j)]
nc, nd = MinNormSolver._min_norm_element_from2(v1v1, v1v2, v2v2)
new_sol_vec = nc*sol_vec + (1-nc)*new_point
change = new_sol_vec - sol_vec
if torch.sum(torch.abs(change)).item() < MinNormSolver.STOP_CRIT:
return sol_vec, nd
sol_vec = new_sol_vec
def find_min_norm_element_FW(vecs):
'''
Given a list of vectors (vecs), this method finds the minimum norm element in the convex hull
as min |u|_2 st. u = \sum c_i vecs[i] and \sum c_i = 1.
It is quite geometric, and the main idea is the fact that if d_{ij} = min |u|_2 st u = c x_i + (1-c) x_j; the solution lies in (0, d_{i,j})
Hence, we find the best 2-task solution, and then run the Frank Wolfe until convergence
'''
# Solution lying at the combination of two points
dps = {}
init_sol, dps = MinNormSolver._min_norm_2d(vecs, dps)
n=len(vecs)
sol_vec = torch.zeros(n)
sol_vec[init_sol[0][0]] = init_sol[1]
sol_vec[init_sol[0][1]] = 1 - init_sol[1]
if n < 3:
# This is optimal for n=2, so return the solution
return sol_vec , init_sol[2]
iter_count = 0
grad_mat = torch.zeros((n,n))
for i in range(n):
for j in range(n):
grad_mat[i,j] = dps[(i, j)]
while iter_count < MinNormSolver.MAX_ITER:
t_iter = torch.argmin(torch.dot(grad_mat, sol_vec))
v1v1 = torch.dot(sol_vec, torch.dot(grad_mat, sol_vec))
v1v2 = torch.dot(sol_vec, grad_mat[:, t_iter])
v2v2 = grad_mat[t_iter, t_iter]
nc, nd = MinNormSolver._min_norm_element_from2(v1v1, v1v2, v2v2)
new_sol_vec = nc*sol_vec
new_sol_vec[t_iter] += 1 - nc
change = new_sol_vec - sol_vec
if torch.sum(torch.abs(change)) < MinNormSolver.STOP_CRIT:
return sol_vec, nd
sol_vec = new_sol_vec
def gradient_normalizers(grads, losses, normalization_type):
gn = {}
if normalization_type == 'l2':
for t in grads:
gn[t] = torch.sqrt(torch.sum([gr.pow(2).sum().data.cpu() for gr in grads[t]]))
elif normalization_type == 'loss':
for t in grads:
gn[t] = losses[t]
elif normalization_type == 'loss+':
for t in grads:
gn[t] = losses[t] * torch.sqrt(torch.sum([gr.pow(2).sum().data.cpu() for gr in grads[t]]))
elif normalization_type == 'none':
for t in grads:
gn[t] = 1.0
else:
print('ERROR: Invalid Normalization Type')
return gn