24 votes
Accepted

Deep Reinforcement Learning for General Purpose Optimization

"General purpose optimization" is quite broad, so I'll take a step back first, to better identifying the motivation of using ML in optimization settings. To keep things simple, I'll consider ...
user avatar
  • 3,253
24 votes
Accepted

What are the tradeoffs between "exact" and Reinforcement Learning methods for solving optimization problems

As far as I understand it, all machine learning approaches used for solving (combinatorial) optimization problems, and in particular reinforcement learning, work as follows: Use a greedy algorithm to ...
user avatar
16 votes
Accepted

What is the connection of Operations Research and Reinforcement Learning?

In the operations research community, many study dynamic programming models and solution approaches. Since many interesting dynamic programming problems attempt to model information uncertainty over ...
user avatar
  • 1,502
14 votes

What are the tradeoffs between "exact" and Reinforcement Learning methods for solving optimization problems

Reinforcement learning is set of the algorithms which are used to solve Markov Decision Processes and its variants, e.g. Partially Observed MDP (POMDP). Most of the problems that we deal with them ...
user avatar
11 votes
Accepted

Online Education for OR and Developing Decision Support Systems

Check Coursera, edX, Udemy, or any other online courses (such as those of Stanford). For example: Free: Discrete Optimization course on Coursera, covers column generation and an introduction to (meta)...
user avatar
  • 5,641
11 votes

Suggestion of some courses in sequential decision making

There are a few courses on Coursera that offer such learning materials. Greedy Algorithms, Minimum Spanning Trees, and Dynamic Programming (Intermediate) The primary topics in this part of the ...
user avatar
  • 5,010
10 votes

What is the connection of Operations Research and Reinforcement Learning?

RL (and Machine Learning in general) often use specialized forms of continuous optimization problems that enable one to use elegant solving techniques (e.g. training/learning on the GPU or even ...
user avatar
  • 2,667
8 votes

What are the tradeoffs between "exact" and Reinforcement Learning methods for solving optimization problems

A main reason to use Reinforced Learning (RL) is because you don't know the dynamics (update rules) of the system. If you don't know the details of how the system will update, you will not be able to ...
user avatar
7 votes

What is the connection of Operations Research and Reinforcement Learning?

Something that I think is becoming more prominent is the use of RL for what would have traditionally been solved using exact solutions. https://medium.com/octavian-ai/finding-shortest-paths-with-...
user avatar
  • 1,059
6 votes

Online Education for OR and Developing Decision Support Systems

For heuristics you can take a look at this course. Also the book "How to solve it : Modern heuristics" by Zbigniew Michalewicz and David B. Fogel. Reinforcement Learning has similiarities with ...
user avatar
  • 2,809
5 votes

Online Education for OR and Developing Decision Support Systems

Institute of Applied Optimization Metaheuristic Optimization - prerequisite: Java programming Coursera Practical Reinforcement Learning - related
user avatar
  • 5,010
4 votes

Online Education for OR and Developing Decision Support Systems

For Reinforcement learning and Deep Reinforcement learning, a wonderful online free course on RL by David Silver (the first author of the AlphaZero, AlphaGo algorithms) exists: YouTube links A more ...
user avatar
4 votes

Suggestion of some courses in sequential decision making

The previous answer provided a great list on the classical dynamic programming. For Reinforcement learning and Deep Reinforcement learning, a wonderful online free course on RL by David Silver (the ...
user avatar
3 votes

What are the tradeoffs between "exact" and Reinforcement Learning methods for solving optimization problems

Reinforcement learning does not seem to be the most obvious choice in machine learning for optimisation. I have heard a lot of using supervised learning to decide the decisions to take when exploring ...
user avatar
  • 938
1 vote
Accepted

Determining which choice strategy (epsilon-greedy, softmax, UCB, etc.) fits participant behavior in a multi armed bandit study

Figured it out. It works exactly they way I originally intended to do it in my question. See this tutorial for detailed information on how to actually implement the procedure with real data.
user avatar
1 vote
Accepted

Understanding MDP's Dual Linear Program

It is just $$\sum_{n=1}^\infty a_n = \sum_{n=0}^\infty a_n - a_0 = \sum_{n=1}^\infty a_{n-1} - a_0,$$ where $$a_n=\lambda^n P^{d^\infty}(X_{n+1}=j\mid X_1=k).$$ In particular $$a_0=\lambda^0 P^{d^\...
user avatar
  • 21.7k
1 vote

Can we use reinforcement learning and convex optimization to solve an optimization problem?

You can use RL in any step. But problem is optimality check of solution which is explained above. Also you can solve your problem directly using RL such as RL for VRP. And you can read this blog ...
user avatar
  • 746
1 vote

Can we use reinforcement learning and convex optimization to solve an optimization problem?

Not really, but approximately. By OR standards, a problem is "solved" once we manage to satisfy the KKT conditions. There is no machine learning algorithm to date that can consistently ...
user avatar
1 vote
Accepted

Can Deep RL be used to find optimal division point in an application?

This is a clustering problem, i.e., you want split your services into two clusters, each of which will run on a different device. The way you would do this with RL is to use the goals an objective ...
user avatar

Only top scored, non community-wiki answers of a minimum length are eligible