deep learning architectures
Frameworks for organizing and implementing deep neural networks, including convolutional networks (CNNs), recurrent networks (RNNs), and transformers, each suited for specific tasks.
- BioOSS: A Bio-Inspired Oscillatory State System with Spatio-Temporal Dynamics
- BlurDM: A Blur Diffusion Model for Image Deblurring
- Causally Reliable Concept Bottleneck Models
- Copresheaf Topological Neural Networks: A Generalized Deep Learning Framework
- Gompertz Linear Units: Leveraging Asymmetry for Enhanced Learning Dynamics
- Identifiability of Deep Polynomial Neural Networks
- Identifiability of Deep Polynomial Neural Networks
- On the Closed-Form of Flow Matching: Generalization Does Not Arise from Target Stochasticity
- On the Closed-Form of Flow Matching: Generalization Does Not Arise from Target Stochasticity
- Scaling and context steer LLMs along the same computational path as the human brain
- Understanding the Evolution of the Neural Tangent Kernel at the Edge of Stability