Mohammad Isa Irawan

Studi Komparatif antara Jaringan Syaraf Tiruan Boltzman Machine dan Algoritma Genetika untuk Optimasi Traveling Salesman Problem

Introduction

Studi komparatif antara jaringan syaraf tiruan boltzman machine dan algoritma genetika untuk optimasi traveling salesman problem. Studi komparatif Jaringan Syaraf Tiruan Boltzman Machine dan Algoritma Genetika untuk optimasi Traveling Salesman Problem (TSP). Temukan metode penyelesaian TSP yang lebih baik.

Abstract

Traveling Salesman Problem (TSP) dikenal sebagai suatu permasalahan optimasi klasik dan Non Deterministic Polynomial-time Complete (NPC). Permasalahan ini melibatkan se- orang salesman yang harus melakukan kunjungan sekali pada semua kota sebelum kembali ke kota awalnya, sampai akhirnya perjalanan itu disebut sempurna. Penyelesaian dari ma- salah ini adalah mencari nilai optimum yang paling murah, misalkan perjalanan dengan jarak terpendek atau yang mempunyai total harga yang termurah. Dalam paper ini akan dianalisis penyelesaian TSP dengan JST Boltzman Machine dan Algoritma Genetika. Dari hasil komparasi tersebut ternyata JST Boltzman Machine mem- berikan hasil lebih baik untuk menyelesaikan masalah TSP. Kata kunci : Jaringan Syaraf Tiruan, Boltzman Machine , Algoritma Genetika, TSP.

Review

The paper titled "Studi Komparatif antara Jaringan Syaraf Tiruan Boltzman Machine dan Algoritma Genetika untuk Optimasi Traveling Salesman Problem" addresses the classic and computationally challenging Traveling Salesman Problem (TSP), a known NP-Complete optimization task. The primary objective of this study is to analyze and compare the efficacy of two prominent metaheuristic approaches, the Boltzmann Machine Neural Network and Genetic Algorithms, in finding optimal or near-optimal solutions for TSP instances. This comparative analysis is a valuable endeavor, aiming to shed light on which of these powerful optimization techniques might be more suitable for this particular problem domain. The abstract clearly outlines the application of Jaringan Syaraf Tiruan (JST) Boltzman Machine and Algoritma Genetika (Genetic Algorithm) to tackle the TSP. Both methods are well-established in the field of computational intelligence for solving complex optimization problems, making their direct comparison highly relevant. The paper's core finding, as stated in the abstract, indicates that the Boltzmann Machine provides superior results for solving the TSP compared to the Genetic Algorithm. This conclusion, if robustly supported within the full paper, could offer significant insights for researchers and practitioners seeking effective heuristics for tour optimization. While the abstract provides a clear comparative outcome, a comprehensive review would benefit from further details in the full paper. It would be valuable to understand the specific experimental setup, including the characteristics and scale (e.g., number of cities) of the TSP instances used for comparison, as the performance of metaheuristics can vary significantly with problem size. Additionally, details regarding the parameter configurations for both the Boltzmann Machine (e.g., annealing schedule, number of neurons) and the Genetic Algorithm (e.g., population size, crossover/mutation rates) are crucial for reproducibility and a deeper understanding of the observed performance difference. Elucidating the *degree* of improvement and potentially comparing computational costs alongside solution quality would further strengthen the claims regarding the Boltzmann Machine's superior performance.

Full Text

Comments

You need to be logged in to post a comment.