arXiv:2511.17687v1 Announce Type: new 
Abstract: The brain's Path Integration (PI) mechanism offers substantial guidance and inspiration for Brain-Inspired Navigation (BIN). However, the PI capability constructed by the Continuous Attractor Neural Networks (CANNs) in most existing BIN studies exhibits significant computational redundancy, and its operational efficiency needs to be improved; otherwise, it will not be conducive to the practicality of BIN technology. To address this, this paper proposes an efficient PI approach using representation learning models to replicate CANN neurodynamic patterns. This method successfully replicates the neurodynamic patterns of CANN-modeled Head Direction Cells (HDCs) and Grid Cells (GCs) using lightweight Artificial Neural Networks (ANNs). These ANN-reconstructed HDC and GC models are then integrated to achieve brain-inspired PI for Dead Reckoning (DR). Benchmark tests in various environments, compared with the well-known NeuroSLAM system, demonstrate that this work not only accurately replicates the neurodynamic patterns of navigation cells but also matches NeuroSLAM in positioning accuracy. Moreover, efficiency improvements of approximately 17.5% on the general-purpose device and 40~50% on the edge device were observed, compared with NeuroSLAM. This work offers a novel implementation strategy to enhance the practicality of BIN technology and holds potential for further extension.

اقترحت دراسة جديدة نهجًا فعالًا لدمج المسار (PI) يستخدم نماذج التعلم التمثيلي لتكرار الأنماط الديناميكية العصبية لشبكات الأعصاب الجاذبة المستمرة (CANNs). تنجح هذه الطريقة في إعادة بناء خلايا اتجاه الرأس (HDCs) وخلايا الشبكة (GCs) باستخدام الشبكات العصبية الاصطناعية (ANNs) الخفيفة، مما يعزز الكفاءة التشغيلية لتكنولوجيا الملاحة المستوحاة من الدماغ (BIN).

Un nuevo estudio ha propuesto un enfoque eficiente de Integración de Trayectoria (PI) que utiliza modelos de aprendizaje de representación para replicar los patrones neurodinámicos de las Redes Neuronales de Atractor Continuo (CANNs). Este método reconstruye con éxito las Células de Dirección de Cabeza (HDCs) y las Células de Rejilla (GCs) utilizando Redes Neuronales Artificiales (ANNs) ligeras, mejorando así la eficiencia operativa de la tecnología de Navegación Inspirada en el Cerebro (BIN).

Une nouvelle étude a proposé une approche efficace d'intégration de chemin (PI) qui utilise des modèles d'apprentissage de représentation pour reproduire les motifs neurodynamiques des Réseaux Neuronaux à Attracteurs Continus (CANNs). Cette méthode reconstruit avec succès les Cellules de Direction de Tête (HDCs) et les Cellules de Grille (GCs) à l'aide de Réseaux Neuronaux Artificiels (ANNs) légers, améliorant ainsi l'efficacité opérationnelle de la technologie de Navigation Inspirée du Cerveau (BIN).

A new study has proposed an efficient Path Integration (PI) approach that utilizes representation learning models to replicate the neurodynamic patterns of Continuous Attractor Neural Networks (CANNs). This method successfully reconstructs Head Direction Cells (HDCs) and Grid Cells (GCs) using lightweight Artificial Neural Networks (ANNs), enhancing the operational efficiency of Brain-Inspired Navigation (BIN) technology.

Boosting Brain-inspired Path Integration Efficiency via Learning-based Replication of Continuous Attractor Neurodynamics

arXiv:2502.05509v2 Announce Type: replace 
Abstract: As machine learning models become integral to security-sensitive applications, concerns over data leakage from adversarial attacks continue to rise. Model Inversion (MI) attacks pose a significant privacy threat by enabling adversaries to reconstruct training data from model outputs. While MI attacks on Artificial Neural Networks (ANNs) have been widely studied, Spiking Neural Networks (SNNs) remain largely unexplored in this context. Due to their event-driven and discrete computations, SNNs introduce fundamental differences in information processing that may offer inherent resistance to such attacks. A critical yet underexplored aspect of this threat lies in black-box settings, where attackers operate through queries without direct access to model parameters or gradients-representing a more realistic adversarial scenario in deployed systems. This work presents the first study of black-box MI attacks on SNNs. We adapt a generative adversarial MI framework to the spiking domain by incorporating rate-based encoding for input transformation and decoding mechanisms for output interpretation. Our results show that SNNs exhibit significantly greater resistance to MI attacks than ANNs, as demonstrated by degraded reconstructions, increased instability in attack convergence, and overall reduced attack effectiveness across multiple evaluation metrics. Further analysis suggests that the discrete and temporally distributed nature of SNN decision boundaries disrupts surrogate modeling, limiting the attacker's ability to approximate the target model.

تم إجراء دراسة حول هجمات عكس النموذج (MI) في إعدادات الصندوق الأسود التي تستهدف الشبكات العصبية النابضة (SNN)، مما يبرز التهديدات المحتملة للخصوصية التي تمثلها هذه الهجمات من خلال السماح للخصوم بإعادة بناء بيانات التدريب من مخرجات النموذج. تمثل هذه الأبحاث خطوة مهمة في فهم نقاط الضعف في SNNs في التطبيقات الحساسة للأمان.

Se ha realizado un estudio sobre los ataques de inversión de modelo (MI) en entornos de caja negra dirigidos a redes neuronales de picos (SNN), destacando las amenazas potenciales a la privacidad que estos ataques representan al permitir que los adversarios reconstruyan datos de entrenamiento a partir de las salidas del modelo. Esta investigación marca un paso significativo en la comprensión de las vulnerabilidades de los SNN en aplicaciones sensibles a la seguridad.

Une étude a été menée sur les attaques d'inversion de modèle (MI) en mode boîte noire ciblant les réseaux neuronaux à impulsion (SNN), mettant en évidence les menaces potentielles pour la vie privée que ces attaques posent en permettant aux adversaires de reconstruire des données d'entraînement à partir des sorties du modèle. Cette recherche constitue une avancée significative dans la compréhension des vulnérabilités des SNN dans des applications sensibles à la sécurité.

A study has been conducted on black-box Model Inversion (MI) attacks targeting Spiking Neural Networks (SNNs), highlighting the potential privacy threats these attacks pose by allowing adversaries to reconstruct training data from model outputs. This research marks a significant step in understanding the vulnerabilities of SNNs in security-sensitive applications.

Do Spikes Protect Privacy? Investigating Black-Box Model Inversion Attacks in Spiking Neural Networks

arXiv:2409.11078v2 Announce Type: replace 
Abstract: Artificial Neural Networks (ANNs) have significantly advanced various fields by effectively recognizing patterns and solving complex problems. Despite these advancements, their interpretability remains a critical challenge, especially in applications where transparency and accountability are essential. To address this, explainable AI (XAI) has made progress in demystifying ANNs, yet interpretability alone is often insufficient. In certain applications, model predictions must align with expert-imposed requirements, sometimes exemplified by partial monotonicity constraints. While monotonic approaches are found in the literature for traditional Multi-layer Perceptrons (MLPs), they still face difficulties in achieving both interpretability and certified partial monotonicity. Recently, the Kolmogorov-Arnold Network (KAN) architecture, based on learnable activation functions parametrized as splines, has been proposed as a more interpretable alternative to MLPs. Building on this, we introduce a novel ANN architecture called MonoKAN, which is based on the KAN architecture and achieves certified partial monotonicity while enhancing interpretability. To achieve this, we employ cubic Hermite splines, which guarantee monotonicity through a set of straightforward conditions. Additionally, by using positive weights in the linear combinations of these splines, we ensure that the network preserves the monotonic relationships between input and output. Our experiments demonstrate that MonoKAN not only enhances interpretability but also improves predictive performance across the majority of benchmarks, outperforming state-of-the-art monotonic MLP approaches.

تم تقديم MonoKAN، وهو شبكة كولموغوروف-أرنولد أحادية الاتجاه معتمدة، لتحسين قابلية تفسير الشبكات العصبية الاصطناعية (ANNs) مع ضمان الامتثال لقيود الأحادية الجزئية. يتناول هذا التطور التحديات المستمرة لتحقيق الشفافية والمساءلة في تطبيقات الذكاء الاصطناعي، خاصةً عندما يجب أن تتماشى تنبؤات النموذج مع متطلبات يحددها الخبراء.

MonoKAN, una Red Kolmogorov-Arnold Monótona Certificada, ha sido presentada para mejorar la interpretabilidad de las Redes Neuronales Artificiales (RNA) mientras asegura el cumplimiento de las restricciones de monotonía parcial. Este desarrollo aborda los desafíos en curso para lograr transparencia y responsabilidad en las aplicaciones de IA, especialmente donde las predicciones del modelo deben cumplir con requisitos definidos por expertos.

MonoKAN, un Réseau Kolmogorov-Arnold Monotone Certifié, a été introduit pour améliorer l'interprétabilité des Réseaux de Neurones Artificiels (RNA) tout en garantissant le respect des contraintes de monotonie partielle. Ce développement répond aux défis persistants liés à la transparence et à la responsabilité dans les applications d'IA, en particulier lorsque les prédictions des modèles doivent répondre à des exigences définies par des experts.

MonoKAN, a Certified Monotonic Kolmogorov-Arnold Network, has been introduced to enhance the interpretability of Artificial Neural Networks (ANNs) while ensuring compliance with partial monotonicity constraints. This development addresses the ongoing challenges in achieving transparency and accountability in AI applications, particularly where model predictions must meet expert-defined requirements.

Boosting Brain-inspired Path Integration Efficiency via Learning-based Replication of Continuous Attractor Neurodynamics

Was this article worth reading? Share it

One More Thing in AI

LCW

AI Flow Chat