analiza aiqbench
This research introduces the Bidirectional Subspace Rotation Algorithm (B-SRA) to enhance the robustness of Bidirectional Associative Memory networks against noise and adversarial attacks. By enforcing orthogonal weight matrices and gradient-pattern alignment, the study demonstrates significantly improved resilience in associative memory architectures without relying on backpropagation gradients. These findings offer a practical pathway for building more reliable neural systems capable withstanding data corruption and security threats.
Robust Bidirectional Associative Memory via Regularization Inspired by the Subspace Rotation Algorithm
arXiv:2511.11902v2 Announce Type: replace Abstract: Bidirectional Associative Memory (BAM) trained with Bidirectional Backpropagation (B-BP) often suffers from poor robustness and high sensitivity to noise and adversarial attacks. To address these issues, we propose a novel gradient-free training algorithm, the Bidirectional Subspace Rotation Algorithm (B-SRA), which significantly improves the robustness and convergence behavior of BAM. Through comprehensive experiments, we identify two key principles -- orthogonal weight matrices (OWM) and gradient-pattern alignment (GPA) -- as central to enhancing the robustness of BAM. Motivated by these findings, we introduce new regularization strategies into B-BP, resulting in models with greatly improved resistance to corruption and adversarial perturbations. We further conduct an ablation study across different training strategies to determine the most robust configuration and evaluate BAM's performance under a variety of attack scenarios and memory capacities, including 50, 100, and 200 associative pairs. Among all methods, the SAME configuration, which integrates both OWM and GPA, achieves the strongest resilience. Overall, our results demonstrate that B-SRA and the proposed regularization strategies lead to substantially more robust associative memories and open new directions for building resilient neural architectures.
Abstract:Bidirectional Associative Memory (BAM) trained with Bidirectional Backpropagation (B-BP) often suffers from poor robustness and high sensitivity to noise and adversarial attacks. To address these issues, we propose a novel gradient-free training algorithm, the Bidirectional Subspace Rotation Algorithm (B-SRA), which significantly improves the robustness and convergence behavior of BAM. Through comprehensive experiments, we identify two key principles -- orthogonal weight matrices (OWM) and gradient-pattern alignment (GPA) -- as central to enhancing the robustness of BAM. Motivated by these findings, we introduce new regularization strategies into B-BP, resulting in models with greatly improved resistance to corruption and adversarial perturbations. We further conduct an ablation study across different training strategies to determine the most robust configuration and evaluate BAM's performance under a variety of attack scenarios and memory capacities, including 50, 100, and 200 associative pairs. Among all methods, the SAME configuration, which integrates both OWM and GPA, achieves the strongest resilience. Overall, our results demonstrate that B-SRA and the proposed regularization strategies lead to substantially more robust associative memories and open new directions for building resilient neural architectures.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2511.11902 [cs.LG]
(or arXiv:2511.11902v2 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2511.11902 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 272 cuvinte
Subiecte conexe
Generalization and infinite width
LessWrong · 1 aug. 2026
Game Design Creativity evals
LessWrong · 12 oct. 2026
Productionizing TurboQuant on AMD GPUs for KV-Cache-Bound LLM Inference
AMD · 11 iun. 2026
Divisive Normalization Shapes Low-Rank Slow Manifolds for Continuous Working Memory
arXiv · 5 aug. 2026