arXiv:2505.13567v2 Announce Type: replace 
Abstract: Recurrent neural networks (RNNs) trained on neuroscience-inspired tasks offer powerful models of brain computation. However, typical training paradigms rely on open-loop, supervised settings, whereas real-world learning unfolds in closed-loop environments. Here, we develop a mathematical theory describing the learning dynamics of linear RNNs trained in closed-loop contexts. We first demonstrate that two otherwise identical RNNs, trained in either closed- or open-loop modes, follow markedly different learning trajectories. To probe this divergence, we analytically characterize the closed-loop case, revealing distinct stages aligned with the evolution of the training loss. Specifically, we show that the learning dynamics of closed-loop RNNs, in contrast to open-loop ones, are governed by an interplay between two competing objectives: short-term policy improvement and long-term stability of the agent-environment interaction. Finally, we apply our framework to a realistic motor control task, highlighting its broader applicability. Taken together, our results underscore the importance of modeling closed-loop dynamics in a biologically plausible setting.

تستكشف دراسة حديثة ديناميات تعلم الشبكات العصبية المتكررة (RNN) في بيئات مغلقة، والتي تعكس بشكل أكثر دقة السيناريوهات الواقعية مقارنة بالتدريب التقليدي في بيئات مفتوحة. هذه الأبحاث مهمة لأنها تعزز فهمنا لكيفية قدرة الشبكات العصبية على نمذجة العمليات الحسابية في الدماغ بشكل أكثر فعالية، مما قد يؤدي إلى تقدم في الذكاء الاصطناعي وتطبيقات علم الأعصاب.

Un estudio reciente explora la dinámica de aprendizaje de las redes neuronales recurrentes (RNN) en entornos de bucle cerrado, que reflejan de manera más precisa los escenarios del mundo real en comparación con el entrenamiento tradicional en bucle abierto. Esta investigación es significativa ya que mejora nuestra comprensión de cómo las RNN pueden modelar el cálculo cerebral de manera más efectiva, lo que podría llevar a avances en inteligencia artificial y aplicaciones en neurociencia.

Une étude récente explore la dynamique d'apprentissage des réseaux de neurones récurrents (RNN) dans des environnements en boucle fermée, qui reflètent plus fidèlement les scénarios du monde réel par rapport à l'entraînement traditionnel en boucle ouverte. Cette recherche est importante car elle améliore notre compréhension de la manière dont les RNN peuvent modéliser le calcul cérébral de manière plus efficace, ce qui pourrait conduire à des avancées dans l'intelligence artificielle et les applications en neurosciences.

A recent study explores the learning dynamics of recurrent neural networks (RNNs) in closed-loop environments, which more accurately reflect real-world scenarios compared to traditional open-loop training. This research is significant as it enhances our understanding of how RNNs can model brain computation more effectively, potentially leading to advancements in artificial intelligence and neuroscience applications.

Learning Dynamics of RNNs in Closed-Loop Environments

arXiv:2512.04189v1 Announce Type: new 
Abstract: Binary Neural Networks (BNNs), which constrain both weights and activations to binary values, offer substantial reductions in computational complexity, memory footprint, and energy consumption. These advantages make them particularly well suited for deployment on resource-constrained devices. However, training BNNs via gradient-based optimization remains challenging due to the discrete nature of their variables. The dominant approach, quantization-aware training, circumvents this issue by employing surrogate gradients. Yet, this method requires maintaining latent full-precision parameters and performing the backward pass with floating-point arithmetic, thereby forfeiting the efficiency of binary operations during training. While alternative approaches based on local learning rules exist, they are unsuitable for global credit assignment and for back-propagating errors in multi-layer architectures. This paper introduces Binary Error Propagation (BEP), the first learning algorithm to establish a principled, discrete analog of the backpropagation chain rule. This mechanism enables error signals, represented as binary vectors, to be propagated backward through multiple layers of a neural network. BEP operates entirely on binary variables, with all forward and backward computations performed using only bitwise operations. Crucially, this makes BEP the first solution to enable end-to-end binary training for recurrent neural network architectures. We validate the effectiveness of BEP on both multi-layer perceptrons and recurrent neural networks, demonstrating gains of up to +6.89% and +10.57% in test accuracy, respectively. The proposed algorithm is released as an open-source repository.

يقدم مقال جديد خوارزمية انتشار الخطأ الثنائي (BEP) المصممة لتحسين تدريب الشبكات العصبية الثنائية (BNNs)، التي تتميز بأوزانها وتفعيلات ثنائية. تتناول هذه الخوارزمية التحديات المتعلقة بالتحسين القائم على التدرجات في الشبكات العصبية الثنائية، التي تعتمد تقليديًا على طرق التدريب الواعية بالتكميم والتي تؤثر على الكفاءة خلال عملية التدريب.

Un nuevo artículo presenta el algoritmo de Propagación de Errores Binarios (BEP) diseñado para mejorar el entrenamiento de las Redes Neuronales Binarias (BNNs), que se caracterizan por sus pesos y activaciones binarias. Este algoritmo aborda los desafíos de la optimización basada en gradientes en las BNNs, que tradicionalmente dependen de métodos de entrenamiento conscientes de la cuantización que comprometen la eficiencia durante el proceso de entrenamiento.

Un nouvel article présente l'algorithme de propagation d'erreur binaire (BEP) conçu pour améliorer l'entraînement des réseaux neuronaux binaires (BNNs), caractérisés par leurs poids et activations binaires. Cet algorithme aborde les défis de l'optimisation basée sur le gradient dans les BNNs, qui s'appuient traditionnellement sur des méthodes d'entraînement conscientes de la quantification qui compromettent l'efficacité pendant le processus d'entraînement.

A new paper presents the Binary Error Propagation (BEP) algorithm designed to enhance the training of Binary Neural Networks (BNNs), which are characterized by their binary weights and activations. This algorithm addresses the challenges of gradient-based optimization in BNNs, which traditionally rely on quantization-aware training methods that compromise efficiency during the training process.

Learning Dynamics of RNNs in Closed-Loop Environments

Was this article worth reading? Share it

LucidQuery AI

CodeSpaced

The Visualizer