Salvato in:
| Autori principali: | , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2507.04659 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917020444917760 |
|---|---|
| author | Jia, Hanzhang Gao, Yi |
| author_facet | Jia, Hanzhang Gao, Yi |
| contents | To address the challenges posed by the heavy reliance of multi-output models on preset probability distributions and embedded prior knowledge in non-injective regression tasks, this paper proposes a cycle consistency-based data-driven training framework. The method jointly optimizes a forward model Φ: X to Y and a backward model Ψ: Y to X, where the cycle consistency loss is defined as L _cycleb equal L(Y reduce Φ(Ψ(Y))) (and vice versa). By minimizing this loss, the framework establishes a closed-loop mechanism integrating generation and validation phases, eliminating the need for manual rule design or prior distribution assumptions. Experiments on normalized synthetic and simulated datasets demonstrate that the proposed method achieves a cycle reconstruction error below 0.003, achieving an improvement of approximately 30% in evaluation metrics compared to baseline models without cycle consistency. Furthermore, the framework supports unsupervised learning and significantly reduces reliance on manual intervention, demonstrating potential advantages in non-injective regression tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_04659 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A Cycle-Consistency Constrained Framework for Dynamic Solution Space Reduction in Noninjective Regression Jia, Hanzhang Gao, Yi Machine Learning To address the challenges posed by the heavy reliance of multi-output models on preset probability distributions and embedded prior knowledge in non-injective regression tasks, this paper proposes a cycle consistency-based data-driven training framework. The method jointly optimizes a forward model Φ: X to Y and a backward model Ψ: Y to X, where the cycle consistency loss is defined as L _cycleb equal L(Y reduce Φ(Ψ(Y))) (and vice versa). By minimizing this loss, the framework establishes a closed-loop mechanism integrating generation and validation phases, eliminating the need for manual rule design or prior distribution assumptions. Experiments on normalized synthetic and simulated datasets demonstrate that the proposed method achieves a cycle reconstruction error below 0.003, achieving an improvement of approximately 30% in evaluation metrics compared to baseline models without cycle consistency. Furthermore, the framework supports unsupervised learning and significantly reduces reliance on manual intervention, demonstrating potential advantages in non-injective regression tasks. |
| title | A Cycle-Consistency Constrained Framework for Dynamic Solution Space Reduction in Noninjective Regression |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2507.04659 |