recurrent neural networks
A class of neural networks specifically designed for sequence data, where connections between nodes can create cycles. They are well-suited for tasks involving time series prediction and natural language processing due to their ability to maintain contextual information.
- Concept-Guided Interpretability via Neural Chunking
- EDELINE: Enhancing Memory in Diffusion-based World Models via Linear-Time Sequence Modeling
- Efficient Allocation of Working Memory Resource for Utility Maximization in Humans and Recurrent Neural Networks
- Finding separatrices of dynamical flows with Deep Koopman Eigenfunctions
- Flow Equivariant Recurrent Neural Networks
- Hardware-aligned Hierarchical Sparse Attention for Efficient Long-term Memory Access
- High-dimensional neuronal activity from low-dimensional latent dynamics: a solvable model
- High-dimensional neuronal activity from low-dimensional latent dynamics: a solvable model
- Learning Dynamics of RNNs in Closed-Loop Environments
- Measuring and Controlling Solution Degeneracy across Task-Trained Recurrent Neural Networks
- Mechanistic Interpretability of RNNs emulating Hidden Markov Models
- Metric Automata Theory: A Unifying Theory of RNNs
- Nonparametric Quantile Regression with ReLU-Activated Recurrent Neural Networks
- RNNs perform task computations by dynamically warping neural representations
- Revisiting Bi-Linear State Transitions in Recurrent Neural Networks
- Revisiting Glorot Initialization for Long-Range Linear Recurrences
- Separating the 'what' and 'how' of compositional computation to enable reuse and continual learning
- Shaping Sequence Attractor Schema in Recurrent Neural Networks
- Slow Transition to Low-Dimensional Chaos in Heavy-Tailed Recurrent Neural Networks
- The Logical Expressiveness of Temporal GNNs via Two-Dimensional Product Logics
- Unfolding the Black Box of Recurrent Neural Networks for Path Integration
- Universal Sequence Preconditioning