Supplementing recurrent neural networks with annealing to solve combinatorial optimization problems

Shoummo Ahsan Khandoker, Jawaril Munshad Abedin, Mohamed Hibat-Allah

Machine Learning Science and Technology 2023 article cited by 14

Source Open-access PDF More in Mathematics

importance 4/10 ai assisted Metaheuristic Optimization Algorithms Research operations research stage: research

Problem

Simulated annealing suffers slow convergence and gets stuck in local search spaces when solving combinatorial optimization problems like Max-Cut, nurse scheduling, and TSP.

Why it matters

Scheduling and routing problems of this type affect real operational efficiency across industries.

What the AI did

The variational classical annealing (VCA) framework combines autoregressive recurrent neural networks with traditional annealing to sample uncorrelated solutions, tested on three benchmark problems.

Results

not stated in source

Impact

Demonstrated potential on three benchmark problems; real-world adoption not stated in source.

Record

Venue	Machine Learning Science and Technology
Published	2023
Work type	article
Authors	Shoummo Ahsan Khandoker, Jawaril Munshad Abedin, Mohamed Hibat-Allah
Citations	14 (OpenAlex, at import time)
References	66
Model or method	VCA (variational classical annealing) with autoregressive RNNs
Organization	not stated in source
Primary source	https://doi.org/10.1088/2632-2153/acb895
Open access	gold
Keywords	combinatorial optimization, simulated annealing, recurrent neural networks, Max-Cut, nurse scheduling, TSP

Importance rating

4 / 10

real but narrower advance

Abstract frames VCA as approach to real-world optimization problems (Max-Cut, nurse scheduling, TSP), outperforming SA by orders of magnitude in relative error; no affected-population scale stated.

Basis: inferred. Judged from the source text, which gives no explicit scale. Stated scale: not stated in source. Method and limits.

More in Mathematics

#	Paper	Year	Citations	Importance
1	Discovering faster matrix multiplication algorithms with reinforcement learning Alhussein Fawzi, Matej Balog, Aja Huang, Thomas Hubert, Bernardino Rom...	2022	476	7
2	Solving olympiad geometry without human demonstrations Trieu H. Trinh, Yuhuai Wu, Quoc Viet Le, He He, Thang Luong	2024	307	7
3	Mathematical discoveries from program search with large language models Bernardino Romera‐Paredes, Mohammadamin Barekatain, Alexander Novikov,...	2023	509	6
4	Superhuman AI for multiplayer poker Noam Brown, Tüomas Sandholm	2019	583	6
5	Efficient generator of mathematical expressions for symbolic regression Sebastian Mežnar, Sašo Džeroski, Ljupčo Todorovski	2023	20	4