Probing Neural TSP Representations for Prescriptive Decision Support
Reuben Narad, Léonard Boussioux, Michael Wagner · 2026
Read Probing Neural TSP Representations for Prescriptive Decision Support by Reuben Narad, Léonard Boussioux, Michael Wagner (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-07216.txt.
Description
Neural routing solvers are fast, but planners need what-if answers: which stop to drop, which connection to keep, which subset to remove when a route fails. Reuben Narad and two co-authors present Prescriptive Probing, which reuses the frozen representations of a trained solver to rank candidate interventions.
Where exhaustive re-solving supplies labels, supervised probes learn from them; where action spaces are too large, a lightweight reinforcement learning head learns the ranking instead. Tests on real-road routing problems show the probes guiding decisions without paying for a fresh solve each time. Planners get decision support from models they already have, at a fraction of the cost.
Catalog record
| CREATOR | Reuben Narad, Léonard Boussioux, Michael Wagner |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 46 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-07216