Moving beyond R's optim function - optimization

I am trying to use R to estimate a multinomial logit model with a manual specification. I have found a few packages that allow you to estimate MNL models here or here.
I've found some other writings on "rolling" your own MLE function here. However, from my digging around - all of these functions and packages rely on the internal optim function.
In my benchmark tests, optim is the bottleneck. Using a simulated dataset with ~16000 observations and 7 parameters, R takes around 90 seconds on my machine. The equivalent model in Biogeme takes ~10 seconds. A colleague who writes his own code in Ox reports around 4 seconds for this same model.
Does anyone have experience with writing their own MLE function or can point me in the direction of something that is optimized beyond the default optim function (no pun intended)?
If anyone wants the R code to recreate the model, let me know - I'll glady provide it. I haven't provided it since it isn't directly relevant to the problem of optimizing the optim function and to preserve space...
EDIT: Thanks to everyone for your thoughts. Based on a myriad of comments below, we were able to get R in the same ballpark as Biogeme for more complicated models, and R was actually faster for several smaller / simpler models that we ran. I think the long term solution to this problem is going to involve writing a separate maximization function that relies on a fortran or C library, but am certainly open to other approaches.

Tried with the nlm() function already? Don't know if it's much faster, but it does improve speed. Also check the options. optim uses a slow algorithm as the default. You can gain a > 5-fold speedup by using the Quasi-Newton algorithm (method="BFGS") instead of the default. If you're not concerned too much about the last digits, you can also set the tolerance levels higher of nlm() to gain extra speed.
f <- function(x) sum((x-1:length(x))^2)
a <- 1:5
system.time(replicate(500,
optim(a,f)
))
user system elapsed
0.78 0.00 0.79
system.time(replicate(500,
optim(a,f,method="BFGS")
))
user system elapsed
0.11 0.00 0.11
system.time(replicate(500,
nlm(f,a)
))
user system elapsed
0.10 0.00 0.09
system.time(replicate(500,
nlm(f,a,steptol=1e-4,gradtol=1e-4)
))
user system elapsed
0.03 0.00 0.03

Did you consider the material on the CRAN Task View for Optimization ?

I am the author of the R package optimParallel, which could be helpful in your case. The package provides parallel versions of the gradient-based optimization methods of optim(). The main function of the package is optimParallel(), which has the same usage and output as optim(). Using optimParallel() can significantly reduce optimization times as illustrated in the following figure (p is the number of paramters).
See https://cran.r-project.org/package=optimParallel and http://arxiv.org/abs/1804.11058 for more information.

FWIW, I've done this in C-ish, using OPTIF9. You'd be hard-pressed to go faster than that. There are plenty of ways for something to go slower, such as by running an interpreter like R.
Added: From the comments, it's clear that OPTIF9 is used as the optimizing engine. That means that most likely the bulk of the time is spent in evaluating the objective function in R. While it is possible that C functions are being used underneath for some of the operations, there still is interpreter overhead. There is a quick way to determine which lines of code and function calls in R are responsible for most of the time, and that is to pause it with the Escape key and examine the stack. If a statement costs X% of time, it is on the stack X% of the time. You may find that there are operations that are not going to C and should be. Any speedup factor you get this way will be preserved when you find a way to parallelize the R execution.

Related

Getting "DUAL_INFEASIBLE" when solving a very simple linear programming problem

I am solving a simple LP problem using Gurobi with dual simplex and presolve. I get the model is unbounded but I couldn't see why such a model is unbounded. Can anyone help to tell me where goes wrong?
I attached the log and also the content in the .mps file.
Thanks very much in advance.
Kind regards,
Hongyu.
The output log and .mps file:
Link to the .mps file: https://studntnu-my.sharepoint.com/:u:/g/personal/hongyuzh_ntnu_no/EV5CBhH2VshForCL-EtPvBUBiFT8uZZkv-DrPtjSFi8PGA?e=VHktwf
Gurobi Optimizer version 9.5.2 build v9.5.2rc0 (mac64[arm])
Thread count: 8 physical cores, 8 logical processors, using up to 8 threads
Optimize a model with 1 rows, 579 columns and 575 nonzeros
Coefficient statistics:
Matrix range [3e-02, 5e+01]
Objective range [7e-01, 5e+01]
Bounds range [0e+00, 0e+00]
RHS range [7e+03, 7e+03]
Iteration Objective Primal Inf. Dual Inf. Time
0 handle free variables 0s
Solved in 0 iterations and 0.00 seconds (0.00 work units)
Unbounded model
The easiest way to debug this is to put a bound on the objective, so the model is no longer unbounded. Then inspect the solution. This is a super easy trick that somehow few people know about.
When we do this with a bound of 100000, we see:
phi = 100000.0000
gamma[11] = -1887.4290
(the rest zero). Indeed we can make gamma[11] as negative as we want to obey R0. Note that gamma[11] is not in the objective.
More advice: It is also useful to write out the LP file of the model and study that carefully. You probably would have caught the error and that would have prevented this post.

SLSQP in ScipyOptimizeDriver only executes one iteration, takes a very long time, then exits

I'm trying to use SLSQP to optimise the angle of attack of an aerofoil to place the stagnation point in a desired location. This is purely as a test case to check that my method for calculating the partials for the stagnation position is valid.
When run with COBYLA, the optimisation converges to the correct alpha (6.04144912) after 47 iterations. When run with SLSQP, it completes one iteration, then hangs for a very long time (10, 20 minutes or more, I didn't time it exactly), and exits with an incorrect value. The output is:
Driver debug print for iter coord: rank0:ScipyOptimize_SLSQP|0
--------------------------------------------------------------
Design Vars
{'alpha': array([0.5])}
Nonlinear constraints
None
Linear constraints
None
Objectives
{'obj_cmp.obj': array([0.00023868])}
Driver debug print for iter coord: rank0:ScipyOptimize_SLSQP|1
--------------------------------------------------------------
Design Vars
{'alpha': array([0.5])}
Nonlinear constraints
None
Linear constraints
None
Objectives
{'obj_cmp.obj': array([0.00023868])}
Optimization terminated successfully. (Exit mode 0)
Current function value: 0.0002386835700364719
Iterations: 1
Function evaluations: 1
Gradient evaluations: 1
Optimization Complete
-----------------------------------
Finished optimisation
Why might SLSQP be misbehaving like this? As far as I can tell, there are no incorrect analytical derivatives when I look at check_partials().
The code is quite long, so I put it on Pastebin here:
core: https://pastebin.com/fKJpnWHp
inviscid: https://pastebin.com/7Cmac5GF
aerofoil coordinates (NACA64-012): https://pastebin.com/UZHXEsr6
You asked two questions whos answers ended up being unrelated to eachother:
Why is the model so slow when you use SLSQP, but fast when you use COBYLA
Why does SLSQP stop after one iteration?
1) Why is SLSQP so slow?
COBYLA is a gradient free method. SLSQP uses gradients. So the solid bet was that slow down happened when SLSQP asked for the derivatives (which COBYLA never did).
Thats where I went to look first. Computing derivatives happens in two steps: a) compute partials for each component and b) solve a linear system with those partials to compute totals. The slow down has to be in one of those two steps.
Since you can run check_partials without too much trouble, step (a) is not likely to be the culprit. So that means step (b) is probably where we need to speed things up.
I ran the summary utility (openmdao summary core.py) on your model and saw this:
============== Problem Summary ============
Groups: 9
Components: 36
Max tree depth: 4
Design variables: 1 Total size: 1
Nonlinear Constraints: 0 Total size: 0
equality: 0 0
inequality: 0 0
Linear Constraints: 0 Total size: 0
equality: 0 0
inequality: 0 0
Objectives: 1 Total size: 1
Input variables: 87 Total size: 1661820
Output variables: 44 Total size: 1169614
Total connections: 87 Total transfer data size: 1661820
Then I generated an N2 of your model and saw this:
So we have an output vector that is 1169614 elements long, which means your linear system is a matrix that is about 1e6x1e6. Thats pretty big, and you are using a DirectSolver to try and compute/store a factorization of it. Thats the source of the slow down. Using DirectSolvers is great for smaller models (rule of thumb, is that the output vector should be less than 10000 elements). For larger ones you need to be more careful and use more advanced linear solvers.
In your case we can see from the N2 that there is no coupling anywhere in your model (nothing in the lower triangle of the N2). Purely feed-forward models like this can use a much simpler and faster LinearRunOnce solver (which is the default if you don't set anything else). So I turned off all DirectSolvers in your model, and the derivatives became effectively instant. Make your N2 look like this instead:
The choice of best linear solver is extremely model dependent. One factor to consider is computational cost, another is numerical robustness. This issue is covered in some detail in Section 5.3 of the OpenMDAO paper, and I won't cover everything here. But very briefly here is a summary of the key considerations.
When just starting out with OpenMDAO, using DirectSolver is both the simplest and usually the fastest option. It is simple because it does not require consideration of your model structure, and it's fast because for small models OpenMDAO can assemble the Jacobian into a dense or sparse matrix and provide that for direct factorization. However, for larger models (or models with very large vectors of outputs), the cost of computing the factorization is prohibitively high. In this case, you need to break the solver structure down more intentionally, and use other linear solvers (sometimes in conjunction with the direct solver--- see Section 5.3 of OpenMDAO paper, and this OpenMDAO doc).
You stated that you wanted to use the DirectSolver to take advantage of the sparse Jacobian storage. That was a good instinct, but the way OpenMDAO is structured this is not a problem either way. We are pretty far down in the weeds now, but since you asked I'll give a short summary explanation. As of OpenMDAO 3.7, only the DirectSolver requires an assembled Jacobian at all (and in fact, it is the linear solver itself that determines this for whatever system it is attached to). All other LinearSolvers work with a DictionaryJacobian (which stores each sub-jac keyed to the [of-var, wrt-var] pair). Each sub-jac can be stored as dense or sparse (depending on how you declared that particular partial derivative). The dictionary Jacobian is effectively a form of a sparse-matrix, though not a traditional one. The key takeaway here is that if you use the LinearRunOnce (or any other solver), then you are getting a memory efficient data storage regardless. It is only the DirectSolver that changes over to a more traditional assembly of an actual matrix object.
Regarding the issue of memory allocation. I borrowed this image from the openmdao docs
2) Why does SLSQP stop after one iteration?
Gradient based optimizations are very sensitive to scaling. I ploted your objective function inside your allowed design space and got this:
So we can see that the minimum is at about 6 degrees, but the objective values are TINY (about 1e-4).
As a general rule of thumb, getting your objective to around order of magnitude 1 is a good idea (we have a scaling report feature that helps with this). I added a reference that was about the order of magnitude of your objective:
p.model.add_objective('obj', ref=1e-4)
Then I got a good result:
Optimization terminated successfully (Exit mode 0)
Current function value: [3.02197589e-11]
Iterations: 7
Function evaluations: 9
Gradient evaluations: 7
Optimization Complete
-----------------------------------
Finished optimization
alpha = [6.04143334]
time: 2.1188600063323975 seconds
Unfortunately, scaling is just hard with gradient based optimization. Starting by scaling your objective/constraints to order-1 is a decent rule of thumb, but its common that you need to adjust things beyond that for more complex problems.

Is there any way to make Z3 use multiple cores (multithreaded) for big problems?

I am attempting to move from a commercial solver to Z3 for large integer satisfiability problem. By "large" I mean that the model I am trying to solve has on the order of 300,000 integers and 300,000 (assert (=... statements, each with a combination of perhaps 8-16 variables.
Our commercial solver took 1353 seconds to solve the big problem. Our commercial solver is actually an optimizer and this was solved as a mixed integer optimization problem. The problem transformed into an integer problem with 5,093,121 variables, 9901 constraints, 63,450,472 zeros, 5,093,120 integers, and it was solved in 4690 iterations. However, it was a simple SAT problem, so I'm hoping to move this to Z3 and ditch the commercial optimizer.
As I indicated, the commercial optimizer took 1353 seconds, but it was also allowed to use 32 cores and indications are that I used many of them (I didn't track how many cores it ended up using).
I would like Z3 to be able to use multiple cores. At the present time it doesn't seem that it does. Is there any way to make it do so? Failing that, is there another SMT solver that will?
Z3 does support parallel processing, see: https://theory.stanford.edu/~nikolaj/programmingz3.html#sec-parallel-z3
Parameters can be set on the command line. So to make z3 use 4 threads and process file solve.z3, use:
z3 parallel.enable=true parallel.threads.max=4 solve.z3
Note that if parallel.enable is set to true, Z3 will default to the number of processors.
Unfortunately this feature is rather poorly documented. Please do report your findings if you try it out!

scipy.sparse.linalg: what's the difference between splu and factorized?

What's the difference between using
scipy.sparse.linalg.factorized(A)
and
scipy.sparse.linalg.splu(A)
Both of them return objects with .solve(rhs) method and for both it's said in the documentation that they use LU decomposition. I'd like to know the difference in performance for both of them.
More specificly, I'm writing a python/numpy/scipy app that implements dynamic FEM model. I need to solve an equation Au = f on each timestep. A is sparse and rather large, but doesn't depend on timestep, so I'd like to invest some time beforehand to make iterations faster (there may be thousands of them). I tried using scipy.sparse.linalg.inv(A), but it threw memory exceptions when the size of matrix was large. I used scipy.linalg.spsolve on each step until recently, and now am thinking on using some sort of decomposition for better performance. So if you have other suggestions aside from LU, feel free to propose!
They should both work well for your problem, assuming that A does not change with each time step.
scipy.sparse.linalg.inv(A) will return a dense matrix that is the same size as A, so it's no wonder it's throwing memory exceptions.
scipy.linalg.solve is also a dense linear solver, which isn't what you want.
Assuming A is sparse, to solve Au=f and you only want to solve Au=f once, you could use scipy.sparse.linalg.spsolve. For example
u = spsolve(A, f)
If you want to speed things up dramatically for subsequent solves, you would instead use scipy.sparse.linalg.factorized or scipy.sparse.linalg.splu. For example
A_inv = splu(A)
for t in range(iterations):
u_t = A_inv.solve(f_t)
or
A_solve = factorized(A)
for t in range(iterations):
u_t = A_solve(f_t)
They should both be comparable in speed, and much faster than the previous options.
As #sascha said, you will need to dig into the documentation to see the differences between splu and factorize. But, you can use 'umfpack' instead of the default 'superLU' if you have it installed and set up correctly. I think umfpack will be faster in most cases. Keep in mind that if your matrix A is too large or has too many non-zeros, an LU decomposition / direct solver may take too much memory on your system. In this case, you might be stuck with using an iterative solver such as this. Unfortunately, you wont be able to reuse the solve of A at each time step, but you might be able to find a good preconditioner for A (approximation to inv(A)) to feed the solver to speed it up.

Time complexity of a genetic algorithm for bin packing

I am trying to explore genetic algorithms (GA) for the bin packing problem, and compare it to classical Any-Fit algorithms. However the time complexity for GA is never mentioned in any of the scholarly articles. Is this because the time complexity is very high? and that the main goal of a GA is to find the best solution without considering the time? What is the time complexity of a basic GA?
Assuming that termination condition is number of iterations and it's constant then in general it would look something like that:
O(p * Cp * O(Crossover) * Mp * O(Mutation) * O(Fitness))
p - population size
Cp - crossover probability
Mp - mutation probability
As you can see it not only depends on parameters like eg. population size but also on implementation of crossover, mutation operations and fitness function implementation. In practice there would be more parameters like for example chromosome size etc.
You don't see much about time complexity in publications because researchers most of the time compare GA using convergence time.
Edit Convergence Time
Every GA has some kind of a termination condition and usually it's convergence criteria. Let's assume that we want to find the minimum of a mathematical function so our convergence criteria will be the function's value. In short we reach convergence during optimization when it's no longer worth it to continue optimization because our best individual doesn't get better significantly. Take a look at this chart:
You can see that after around 10000 iterations fitness doesn't improve much and the line is getting flat. Best case scenario reaches convergence at around 9500 iterations, after that point we don't observe any improvement or it's insignificantly small. Assuming that each line shows different GA then Best case has the best convergence time becuase it reaches convergence criteria first.