Fishers
Fast Approximate Natural Gradient Descent in a Kronecker Factored Eigenbasis
Thomas George, César Laurent, Xavier Bouthillier, Nicolas Ballas, Pascal Vincent
Neural Information Processing Systems http://nips.cc/
Country:
- North America > Canada > Quebec > Montreal (0.04)
- Oceania > Tonga (0.04)
- North America > United States > Indiana > Hamilton County > Fishers (0.04)
Technology:
Country:
- North America > United States > North Carolina (0.04)
- North America > United States > Indiana > Hamilton County > Fishers (0.04)
Technology:
Country:
- North America > United States > Indiana > Hamilton County > Fishers (0.04)
- Africa > Senegal > Kolda Region > Kolda (0.04)
Technology:
Country:
- North America > United States > Louisiana > Orleans Parish > New Orleans (0.04)
- North America > Dominican Republic (0.04)
- Europe > United Kingdom > England > Cambridgeshire > Cambridge (0.04)
- (10 more...)
Kronecker-Factored Approximate Curvature for Modern Neural Network Architectures
However, there is currently no framework to apply it to generic architectures, specifically ones with linear weight-sharing layers.
Country:
- Europe > Germany > Baden-Württemberg > Tübingen Region > Tübingen (0.04)
- North America > United States > Indiana > Hamilton County > Fishers (0.04)
- Europe > United Kingdom > England > Cambridgeshire > Cambridge (0.04)
- Europe > Switzerland > Zürich > Zürich (0.04)
Technology: Information Technology > Artificial Intelligence > Machine Learning > Neural Networks > Deep Learning (1.00)
Stein Variational Gradient Descentwith Matrix-Valued Kernels
See Liu & Wang (2016, 2018); Liu (2017) formoredetails.
Country:
- North America > United States > Indiana > Hamilton County > Fishers (0.04)
- North America > Canada > British Columbia > Metro Vancouver Regional District > Vancouver (0.04)
- Europe > United Kingdom > England > Cambridgeshire > Cambridge (0.04)
- Asia > Middle East > Jordan (0.04)
Technology: Information Technology > Artificial Intelligence > Machine Learning > Statistical Learning (0.68)
Country:
- Europe > Sweden > Stockholm > Stockholm (0.05)
- North America > United States > Indiana > Hamilton County > Fishers (0.04)
- Europe > Germany > Baden-Württemberg > Tübingen Region > Tübingen (0.04)
- (12 more...)
Technology:
WoodFisher: EfficientSecond-OrderApproximation forNeuralNetworkCompression
Recently, there has been significant interest in utilizing this information in the context of deep neural networks; however,relatively little isknown about the quality ofexisting approximationsinthiscontext.
Country:
- North America > United States > Wisconsin > Dane County > Madison (0.04)
- North America > United States > Indiana > Hamilton County > Fishers (0.04)
- North America > Canada > British Columbia > Metro Vancouver Regional District > Vancouver (0.04)
- Europe > Switzerland (0.04)
Technology: Information Technology > Artificial Intelligence > Machine Learning > Neural Networks > Deep Learning (0.48)
WoodFisher: EfficientSecond-OrderApproximation forNeuralNetworkCompression
Recently, there has been significant interest in utilizing this information in the context of deep neural networks; however,relatively little isknown about the quality ofexisting approximationsinthiscontext.
Country:
- North America > United States > Wisconsin > Dane County > Madison (0.04)
- North America > United States > Indiana > Hamilton County > Fishers (0.04)
- North America > Canada > British Columbia > Metro Vancouver Regional District > Vancouver (0.04)
- Europe > Switzerland (0.04)
Technology: Information Technology > Artificial Intelligence > Machine Learning > Neural Networks > Deep Learning (0.34)
Country:
- North America > United States > Pennsylvania > Allegheny County > Pittsburgh (0.04)
- North America > United States > Indiana > Hamilton County > Fishers (0.04)
- Europe > United Kingdom > England > Cambridgeshire > Cambridge (0.04)
Technology:
- Information Technology > Artificial Intelligence > Machine Learning (1.00)
- Information Technology > Communications (0.96)
- Information Technology > Data Science (0.93)