HyperNetwork

Introduced by Ha et al. in HyperNetworks

A HyperNetwork is a network that generates a network 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.

Source: HyperNetworks

Latest Papers

PAPER DATE
Controllable Pareto Multi-Task Learning
Xi LinZhiyuan YangQingfu ZhangSam Kwong
2020-10-13
Learning the Pareto Front with Hypernetworks
| Aviv NavonAviv ShamsianGal ChechikEthan Fetaya
2020-10-08
HyperGrid: Efficient Multi-Task Transformers with Grid-wise Decomposable Hyper Projections
Yi TayZhe ZhaoDara BahriDonald MetzlerDa-Cheng Juan
2020-07-12
Learning Implicit Credit Assignment for Multi-Agent Actor-Critic
Meng ZhouZiyu LiuPengwei SuiYixuan LiYuk Ying Chung
2020-07-06
HyperFlow: Representing 3D Objects as Surfaces
Przemysław SpurekMaciej ZiębaJacek TaborTomasz Trzciński
2020-06-15
Hypernetwork-Based Augmentation
Chih-Yang ChenChe-Han ChangEdward Y. Chang
2020-06-11
MetaInv-Net: Meta Inversion Network for Sparse View CT Image Reconstruction
Haimiao ZhangBaodong LiuHengyong YuBin Dong
2020-05-30
On Infinite-Width Hypernetworks
Etai LittwinTomer GalantiLior WolfGreg Yang
2020-03-27
ADWPNAS: Architecture-Driven Weight Prediction for Neural Architecture Search
XuZhangChenjunZhouBoGu
2020-03-03
Comparing the Parameter Complexity of Hypernetworks and the Embedding-Based Alternative
Tomer GalantiLior Wolf
2020-02-23
ISBNet: Instance-aware Selective Branching Networks
Shaofeng CaiYao ShuWei WangGang ChenBeng Chin Ooi
2020-01-01
Electric Analog Circuit Design with Hypernetworks and a Differential Simulator
Michael RotmanLior Wolf
2019-11-08
Hyper-Graph-Network Decoders for Block Codes
Eliya NachmaniLior Wolf
2019-09-05
Blow: a single-scale hyperconditioned flow for non-parallel raw-audio voice conversion
| Joan SerràSantiago PascualCarlos Segura
2019-06-03
Continual learning with hypernetworks
| Johannes von OswaldChristian HenningJoão SacramentoBenjamin F. Grewe
2019-06-03
A Generative Model for Sampling High-Performance and Diverse Weights for Neural Networks
Lior DeutschErik NijkampYu Yang
2019-05-07
Hypernetwork functional image representation
Sylwester KlocekŁukasz MaziarkaMaciej WołczykJacek TaborJakub NowakMarek Śmieja
2019-02-27
Graph HyperNetworks for Neural Architecture Search
Chris ZhangMengye RenRaquel Urtasun
2018-10-12
Hypernetwork Knowledge Graph Embeddings
| Ivana BalaževićCarl AllenTimothy M. Hospedales
2018-08-21
Learning to Compose over Tree Structures via POS Tags
Gehui ShenZhi-Hong DengTing HuangXi Chen
2018-08-18
Generating Neural Networks with Neural Networks
Lior Deutsch
2018-01-06
Stochastic Maximum Likelihood Optimization via Hypernetworks
Abdul-Saboor SheikhKashif RasulAndreas MerentitisUrs Bergmann
2017-12-04
HyperNetworks with statistical filtering for defending adversarial examples
Zhun SunMete OzayTakayuki Okatani
2017-11-06
Bayesian Hypernetworks
David KruegerChin-Wei HuangRiashat IslamRyan TurnerAlexandre LacosteAaron Courville
2017-10-13
HyperNetworks
| David HaAndrew DaiQuoc V. Le
2016-09-27

Components

COMPONENT TYPE
🤖 No Components Found You can add them if they exist; e.g. Mask R-CNN uses RoIAlign

Categories