Description
A HyperNetwork is a network that generates weights for a main network. The behavior of the main network is the same with any usual neural network: it learns to map some raw inputs to their desired targets; whereas the hypernetwork takes a set of inputs that contain information about the structure of the weights and generates the weight for that layer.
Papers Using This Method
Acquiring and Adapting Priors for Novel Tasks via Neural Meta-Architectures2025-07-07Beyond Static Models: Hypernetworks for Adaptive and Generalizable Forecasting in Complex Parametric Dynamical Systems2025-06-24Learning Personalized Utility Functions for Drivers in Ride-hailing Systems Using Ensemble Hypernetworks2025-06-21LFR-PINO: A Layered Fourier Reduced Physics-Informed Neural Operator for Parametric PDEs2025-06-21Generalizable Agent Modeling for Agent Collaboration-Competition Adaptation with Multi-Retrieval and Dynamic Generation2025-06-20Leveraging External Factors in Household-Level Electrical Consumption Forecasting using Hypernetworks2025-06-17PropMEND: Hypernetworks for Knowledge Propagation in LLMs2025-06-10SHIELD: Secure Hypernetworks for Incremental Expansion Learning Defense2025-06-09ExARNN: An Environment-Driven Adaptive RNN for Learning Non-Stationary Power Dynamics2025-05-23Neural Conditional Transport Maps2025-05-21Editing Across Languages: A Survey of Multilingual Knowledge Editing2025-05-20Relation-Aware Graph Foundation Model2025-05-17DISCO: learning to DISCover an evolution Operator for multi-physics-agnostic prediction2025-04-28Uncertainty quantification of neural network models of evolving processes via Langevin sampling2025-04-21Dynamic Allocation Hypernetwork with Adaptive Model Recalibration for Federated Continual Learning2025-03-25Continual Reinforcement Learning for HVAC Systems Control: Integrating Hypernetworks and Transfer Learning2025-03-24Dynamic Allocation Hypernetwork with Adaptive Model Recalibration for FCL2025-03-23HyperNVD: Accelerating Neural Video Decomposition via Hypernetworks2025-03-21HyperDAS: Towards Automating Mechanistic Interpretability with Hypernetworks2025-03-13HyDA: Hypernetworks for Test Time Domain Adaptation in Medical Imaging Analysis2025-03-06