TechPrufer
Sign inGet started

Blogs›Shopify›AI for Traveling Salesperson Problem

AI for Traveling Salesperson Problem

AI for Traveling Salesperson Problem

1
posts
2021

This feature thread tracks the development and application of AI and optimization techniques to solve complex routing problems like the Traveling Salesperson Problem (TSP). Initial efforts focused on addressing computationally hard problems and handling uncertainty in real-world scenarios. This post details Shopify's participation and victory in the AI4TSP Competition, showcasing the use of dialectic search for expected-value tours and deep reinforcement learning for on-the-fly tour construction. The work is relevant to optimizing Shopify's fulfillment system.

2021

Winning AI4TSP: Solving the Travelling Salesperson Problem with Self-programming Machines - Shopify

10/6/2021

This post details Shopify's participation and victory in the AI4TSP Competition. It explains the Traveling Salesperson Problem (TSP) and its complexities, especially under uncertainty. The team employed two distinct heuristic search approaches: dialectic search for Track 1 (predictive tours) and deep reinforcement learning for Track 2 (dynamic tours). The post highlights the concept of self-programming machines and the importance of learning from experience in AI-driven optimization.