Related
Suppose there are three variables that take on discrete integer values, say w1 = {1,2,3,4,5,6,7,8,9,10,11,12}, w2 = {1,2,3,4,5,6,7,8,9,10,11,12}, and w3 = {1,2,3,4,5,6,7,8,9,10,11,12}. The task is to pick one value from each set such that the resulting triplet minimizes some (black box, computationally expensive) cost function.
I've tried the surrogate optimization in Matlab but I'm not sure it is appropriate. I've also heard about simulated annealing but found no implementation applied to this instance.
Which algorithm, apart from exhaustive search, can solve this combinatorial optimization problem?
Any help would be much appreciated.
The requirement/benefit of Simulated Annealing (SA), is that the objective surface is somewhat smooth, that is, we can be close to a solution.
For a completely random spiky surface- you might as well do a random search
If it is anything smooth, or even sometimes, it makes sense to try SA.
The idea is that (sometimes) changing only 1 of the 3 values, we have little effect on out blackbox function.
Here is a basic example to do this with Simulated Annealing, using frigidum in Python
import numpy as np
w1 = np.array( [1,2,3,4,5,6,7,8,9,10,11,12] )
w2 = np.array( [1,2,3,4,5,6,7,8,9,10,11,12] )
w3 = np.array( [1,2,3,4,5,6,7,8,9,10,11,12] )
W = np.array([w1,w2,w3])
LENGTH = 12
I define a black-box using the Rastrigin function.
def rastrigin_function_n( x ):
"""
N-dimensional Rastrigin
https://en.wikipedia.org/wiki/Rastrigin_function
x_i is in [-5.12, 5.12]
"""
A = 10
n = x.shape[0]
return A*n + np.sum( x**2- A*np.cos(2*np.pi * x) )
def black_box( x ):
"""
Transform from domain [1,12] to [-5,5]
to be able to push to rastrigin
"""
x = (x - 6.5) * (5/5.5)
return rastrigin_function_n(x)
Simulated Annealing needs to modify state X. Instead of taking/modifying values directly, we keep track of indices. This simplifies creating new proposals as an index is always an integer we can simply add/subtract 1 modulo LENGTH.
def random_start():
"""
returns 3 random indices
"""
return np.random.randint(0, LENGTH, size=3)
def random_small_step(x):
"""
change only 1 index
"""
d = np.array( [1,0,0] )
if np.random.random() < .5:
d = np.array( [-1,0,0] )
np.random.shuffle(d)
return (x+d) % LENGTH
def random_big_step(x):
"""
change 2 indici
"""
d = np.array( [1,-1,0] )
np.random.shuffle(d)
return (x+d) % LENGTH
def obj(x):
"""
We have a triplet of indici,
1. Calculate corresponding values in W = [w1,w2,w3]
2. Push the values in out black-box function
"""
indices = x
values = W[np.array([0,1,2]), indices]
return black_box(values)
And throw a SA Scheme at it
import frigidum
local_opt = frigidum.sa(random_start=random_start,
neighbours=[random_small_step, random_big_step],
objective_function=obj,
T_start=10**4,
T_stop=0.000001,
repeats=10**3,
copy_state=frigidum.annealing.naked)
I am not sure what the minimum for this function should be, but it found a objective with 47.9095 with indicis np.array([9, 2, 2])
Edit:
For frigidum to change the cooling schedule, use alpha=.9. My experience is that all the work of experiment which cooling scheme works best doesn't out-weight simply let it run a little longer. The multiplication you proposed, (sometimes called geometric) is the standard one, also implemented in frigidum. So to implement Tn+1 = 0.9*Tn you need a alpha=.9. Be aware this cooling step is done after N repeats, so if repeats=100, it will first do 100 proposals before lowering the temperature with factor alpha
Simple variations on current state often works best. Since its best practice to set the initial temperature high enough to make most proposals (>90%) accepted, it doesn't matter the steps are small. But if you fear its soo small, try 2 or 3 variations. Frigidum accepts a list of proposal functions, and combinations can enforce each other.
I have no experience with MINLP. But even if, so many times experiments can surprise us. So if time/cost is small to bring another competitor to the table, yes!
Try every possible combination of the three values and see which has the lowest cost.
I'd like to write a LP problem in the standard format with MatOptInterface, e.i.:
min c'*x
S.t A*x .== b
x >= 0
Now, how can one write this problem with MathOptInterface? I'm having many issues, one of them is how to define the variable "model". For example, if I try to run:
x = add_variables(model,3)
I first would need to declare this model variable. But I don't know how one is supposed to do this on MathOptInterface.
IIUC in your situation model has to be an argument to be specified by the user of your function.
The user can then pass GLPK.Optimizer(), Tulip.Optimizer() or any other optimizer inheriting from MathOptInterface.AbstractOptimizer.
See e.g. Manual#A complete example.
Alternatively you can look at MOI.Utilities.Model but I don't know how to get an optimizer to solve that model.
Here is how to implement the LP solver for standard Simplex format:
function SolveLP(c,A,b,model::MOI.ModelLike)
x = MOI.add_variables(model, length(c));
MOI.set(model, MOI.ObjectiveFunction{MOI.ScalarAffineFunction{Float64}}(),
MOI.ScalarAffineFunction(MOI.ScalarAffineTerm.(c, x), 0.0))
MOI.set(model, MOI.ObjectiveSense(), MOI.MIN_SENSE)
for xi in x
MOI.add_constraint(model, MOI.SingleVariable(xi), MOI.GreaterThan(0.0))
end
for (i,row) in enumerate(eachrow(A))
row_function = MOI.ScalarAffineFunction(MOI.ScalarAffineTerm.(row, x), 0.0);
MOI.add_constraint(model, row_function, MOI.EqualTo(b[i]))
end
MOI.optimize!(model)
p = MOI.get(model, MOI.VariablePrimal(), x);
return p
end
For the model, just choose something like GLPK.Optimizer()
I am trying to use Pytorch for non-convex optimisation, trying to maximise my objective (so minimise in SGD). I would like to bound my dependent variable x > 0, and also have the sum of my x values be less than 1000.
I think I have the penalty implemented correctly in the form of a ramp penalty, but am struggling with the bounding of the x variable. In Pytorch you can set the bounds using clamp but it doesn't seem appropriate in this case. I think this is because optim needs the gradients free under the hood. Full working example:
import torch
from torch.autograd import Variable
import numpy as np
def objective(x, a, b, c): # Want to maximise this quantity (so minimise in SGD)
d = 1 / (1 + torch.exp(-a * (x)))
# Checking constraint
exceeded_limit = constraint(x).item()
#print(exceeded_limit)
obj = torch.sum(d * (b * c - x))
# If overlimit add ramp penalty
if exceeded_limit < 0:
obj = obj - (exceeded_limit * 10)
print("Exceeded limit")
return - obj
def constraint(x, limit = 1000): # Must be > 0
return limit - x.sum()
N = 1000
# x is variable to optimise for
x = Variable(torch.Tensor([1 for ii in range(N)]), requires_grad=True)
a = Variable(torch.Tensor(np.random.uniform(0,100,N)), requires_grad=True)
b = Variable(torch.Tensor(np.random.rand(N)), requires_grad=True)
c = Variable(torch.Tensor(np.random.rand(N)), requires_grad=True)
# Would like to include the clamp
# x = torch.clamp(x, min=0)
# Non-convex methodf
opt = torch.optim.SGD([x], lr=.01)
for i in range(10000):
# Zeroing gradients
opt.zero_grad()
# Evaluating the objective
obj = objective(x, a, b, c)
# Calculate gradients
obj.backward()
opt.step()
if i%1000==0: print("Objective: %.1f" % -obj.item())
print("\nObjective: {}".format(-obj))
print("Limit: {}".format(constraint(x).item()))
if torch.sum(x<0) > 0: print("Bounds not met")
if constraint(x).item() < 0: print("Constraint not met")
Any suggestions as to how to impose the bounds would be appreciated, either using clamp or otherwise. Or generally advice on non-convex optimisation using Pytorch. This is a much simpler and scaled down version of the problem I'm working so am trying to find a lightweight solution if possible. I am considering using a workaround such as transforming the x variable using an exponential function but then you'd have to scale the function to avoid the positive values becoming infinite, and I want some flexibility with being able to set the constraint.
I meet the same problem with you.
I want to apply bounds on a variable in PyTorch, too.
And I solved this problem by the below Way3.
Your example is a little compliex but I am still learning English.
So I give a simpler example below.
For example, there is a trainable variable v, its bounds is (-1, 1)
v = torch.tensor((0.5, ）, require_grad=True)
v_loss = xxxx
optimizer.zero_grad()
v_loss.backward()
optimizer.step()
Way1. RuntimeError: a leaf Variable that requires grad has been used in an in-place operation.
v.clamp_(-1, 1)
Way2. RuntimeError: Trying to backward through the graph a second time, but the buffers have already been freed.
v = torch.clamp(v, -1, +1) # equal to v = v.clamp(-1, +1)
Way3. NotError. I solved this problem in Way3.
with torch.no_grad():
v[:] = v.clamp(-1, +1) # You must use v[:]=xxx instead of v=xxx
I am new to PK modelling and pymc3, but I have been playing around with pymc3 and trying to implement a simple PK model as part of my own learning. Specifically a model that captures this relationship...
Where C(t)(Cpred) is concentration at time t, Dose is the dose given, V is Volume of distribution, CL is clearance.
I have generated some test data (30 subjects) with values of CL =2 , V=10, for 3 doses 100,200,300, and generated data at timepoints 0,1,2,4,8,12, and also included some random error on CL (normal distribution, 0 mean, omega =0.6) and on the residual unexplained error DV = Cpred + sigma, where sigma is normally distributed the SD =0.33. In addition I have included a transformation on C and V with respect to the weight (uniform distribution 50-90) CLi = CL * WT/70; Vi = V * WT/70.
# Create Data for modelling
np.random.seed(0)
# Subject ID's
data = pd.DataFrame(np.arange(1,31), columns=['subject'])
# Dose
Data['dose'] = np.array([100,100,100,100,100,100,100,100,100,100,
200,200,200,200,200,200,200,200,200,200,
300,300,300,300,300,300,300,300,300,300])
# Random Body Weight
data['WT'] = np.random.randint(50,100, size =30)
# Fixed Clearance and Volume for the population
data['CLpop'] =2
data['Vpop']=10
# Error rate for individual clearance rate
OMEGA = 0.66
# Individual clearance rate as a function of weight and omega
data['CLi'] = data['CLpop']*(data['WT']/70)+ np.random.normal(0, OMEGA )
# Individual Volume as a function of weight
data['Vi'] = data['Vpop']*(data['WT']/70)
# Expand dataframe to account for time points
data = pd.concat([data]*6,ignore_index=True )
data = data.sort('subject')
# Add in time points
data['time'] = np.tile(np.array([0,1,2,4,8,12]), 30)
# Create concentration values using equation
data['Cpred'] = data['dose']/data['Vi'] *np.exp(-1*data['CLi']/data['Vi']*data['time'])
# Error rate for DV
SIGMA = 0.33
# Create Dependenet Variable from Cpred + error
data['DV']= data['Cpred'] + np.random.normal(0, SIGMA )
# Create new df with only data for modelling...
df = data[['subject','dose','WT', 'time', 'DV']]
Create arrays ready for model...
# Prepare data from df to model specific arrays
time = np.array(df['time'])
dose = np.array(df['dose'])
DV = np.array(df['DV'])
WT = np.array(df['WT'])
n_patients = len(data['subject'].unique())
subject = data['subject'].values-1
I have built a simple model in pymc3 ....
pk_model = Model()
with pk_model:
# Hyperparameter Priors
sigma = Lognormal('sigma', mu =0, tau=0.01)
V = Lognormal('V', mu =2, tau=0.01)
CL = Lognormal('CL', mu =1, tau=0.01)
# Transformation wrt to weight
CLi = CL*(WT)/70
Vi = V*(WT)/70
# Expected value of outcome
pred = dose/Vi*np.exp(-1*(CLi/Vi)*time)
# Likelihood (sampling distribution) of observations
conc = Normal('conc', mu =pred, tau=sigma, observed = DV)
My expectation was that I should have been able to resolve from the data the constants and error rates that were originally used to generate the data, although I have not been able to do this, although I can get close. In this example...
data['CLi'].mean()
> 2.322473543135788
data['Vi'].mean()
> 10.147619047619049
And the trace shows....
So my questions are..
Is my code structured correctly and are there any glaring mistakes that I have overlooked that might account for this difference?
Can I structure the pymc3 model to better reflect the relationship from which I have generated the data?
What would be your suggestions to improve the model?
Thanks in advance!
I'm going to answer my own question!
But I implemented a hierarchal model following the example found here...
GLM -hierarchical
and it works a treat. Also I noticed errors in the way I was applying the errors in the dataframe - should use
data['CLer'] = np.random.normal(scale=OMEGA, size=30)
To ensure each subject has a different value for the error
I am trying to follow the tutorial of using the optimization tool box in MATLAB. Specifically, I have a function
f = exp(x(1))*(4*x(1)^2+2*x(2)^2+4*x(1)*x(2)+2*x(2)+1)+b
subject to the constraint:
(x(1))^2+x(2)-1=0,
-x(1)*x(2)-10<=0.
and I want to minimize this function for a range of b=[0,20]. (That is, I want to minimize this function for b=0, b=1,b=2 ... and so on).
Below is the steps taken from the MATLAB's tutorial webpage(http://www.mathworks.com/help/optim/ug/nonlinear-equality-and-inequality-constraints.html), how should I change the code so that, the optimization will run for 20 times, and save the optimal values for each b?
Step 1: Write a file objfun.m.
function f = objfun(x)
f = exp(x(1))*(4*x(1)^2+2*x(2)^2+4*x(1)*x(2)+2*x(2)+1)+b;
Step 2: Write a file confuneq.m for the nonlinear constraints.
function [c, ceq] = confuneq(x)
% Nonlinear inequality constraints
c = -x(1)*x(2) - 10;
% Nonlinear equality constraints
ceq = x(1)^2 + x(2) - 1;
Step 3: Invoke constrained optimization routine.
x0 = [-1,1]; % Make a starting guess at the solution
options = optimoptions(#fmincon,'Algorithm','sqp');
[x,fval] = fmincon(#objfun,x0,[],[],[],[],[],[],...
#confuneq,options);
After 21 function evaluations, the solution produced is
x, fval
x =
-0.7529 0.4332
fval =
1.5093
Update:
I tried your answer, but I am encountering problem with your step 2. Bascially, I just fill the my step 2 to your step 2 (below the comment "optimization just like before").
%initialize list of targets
b = 0:1:20;
%preallocate/initialize result vectors using zeros (increases speed)
opt_x = zeros(length(b));
opt_fval = zeros(length(b));
>> for idx = 1, length(b)
objfun = #(x)objfun_builder(x,b)
%optimization just like before
x0 = [-1,1]; % Make a starting guess at the solution
options = optimoptions(#fmincon,'Algorithm','sqp');
[x,fval] = fmincon(#objfun,x0,[],[],[],[],[],[],...
#confuneq,options);
%end the stuff I fill in
opt_x(idx) = x
opt_fval(idx) = fval
end
However, it gave me the output is:
Error: "objfun" was previously used as a variable, conflicting
with its use here as the name of a function or command.
See "How MATLAB Recognizes Command Syntax" in the MATLAB
documentation for details.
There are two things you need to change about your code:
Creation of the objective function.
Multiple optimizations using a loop.
1st Step
For more flexibility with regard to b, you need to set up another function that returns a handle to the desired objective function, e.g.
function h = objfun_builder(x, b)
h = #(x)(objfun(x));
function f = objfun(x)
f = exp(x(1))*(4*x(1)^2+2*x(2)^2+4*x(1)*x(2)+2*x(2)+1) + b;
end
end
A more elegant and shorter approach are anonymous functions, e.g.
objfun_builder = #(x,b)(exp(x(1))*(4*x(1)^2+2*x(2)^2+4*x(1)*x(2)+2*x(2)+1) + b);
After all, this works out to be the same as above. It might be less intuitive for a Matlab-beginner, though.
2nd Step
Instead of placing an .m-file objfun.m in your path, you will need to call
objfun = #(x)(objfun_builder(x,myB));
to create an objective function in your workspace. In order to loop over the interval b=[0,20], use the following loop
%initialize list of targets
b = 0:1:20;
%preallocate/initialize result vectors using zeros (increases speed)
opt_x = zeros(length(b))
opt_fval = zeros(length(b))
%start optimization of list of targets (`b`s)
for idx = 1, length(b)
objfun = #(x)objfun_builder(x,b)
%optimization just like before
opt_x(idx) = x
opt_fval(idx) = fval
end