Abstract:Recent advances in AI have been primarily driven by large-scale neural architectures that excel at function approximation, rather than by tailored inductive biases and inference or learning strategies that could be important for resource-efficient real-world perception and planning through the solution of inverse problems. In this work, we consider the possibility of enabling robust inversion of continuous forward processes $p \mapsto y$ by learning representations of $y$ that are bijectively aligned with $p$ while remaining insensitive to perturbations in $y$ caused by noise or model mismatch. We propose Twincher, a class of architectures based on stacks of structured diffeomorphic transformations and tailored adversarial training strategies that enable learning such bijective representations. We provide a public API for training and inference and empirically demonstrate the ability of the proposed architecture to efficiently learn bijective representations of synthetic systems, thereby enabling robust and efficient iterative inverse inference. Compared to a baseline inverse-modeling approach, the method exhibits improved data efficiency and robustness, providing initial evidence for the potential of bijective representation learning in robotics, vision, and physical AI.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2605.13470 [cs.LG] |
| (or arXiv:2605.13470v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2605.13470 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Arkady Gonoskov [view email]
[v1]
Wed, 13 May 2026 12:57:17 UTC (2,954 KB)