3 papers
cs.LG2024
Adaptive Feedforward Gradient Estimation in Neural ODEs
Jaouad Dabounou
Neural Ordinary Differential Equations (Neural ODEs) represent a significant breakthrough in deep learning, promising to bridge the gap between machine learning and the rich theore…
cs.LG2024
Adaptive Class Emergence Training: Enhancing Neural Network Stability and Generalization through Progressive Target Evolution
Jaouad Dabounou
Recent advancements in artificial intelligence, particularly deep neural networks, have pushed the boundaries of what is achievable in complex tasks. Traditional methods for traini…
cs.LG2024
Enhancing Neural Network Interpretability Through Conductance-Based Information Plane Analysis
Jaouad Dabounou, Amine Baazzouz
The Information Plane is a conceptual framework used to analyze the flow of information in neural networks, but traditional methods based on activations may not fully capture the d…