Picture for Pedro Savarese

Pedro Savarese

Approaching Deep Learning through the Spectral Dynamics of Weights

Add code
Aug 21, 2024
Viaarxiv icon

SySMOL: A Hardware-software Co-design Framework for Ultra-Low and Fine-Grained Mixed-Precision Neural Networks

Add code
Nov 23, 2023
Viaarxiv icon

Accelerated Training via Incrementally Growing Neural Networks using Variance Transfer and Learning Rate Adaptation

Add code
Jun 22, 2023
Viaarxiv icon

Information-Theoretic Segmentation by Inpainting Error Maximization

Add code
Dec 14, 2020
Figure 1 for Information-Theoretic Segmentation by Inpainting Error Maximization
Figure 2 for Information-Theoretic Segmentation by Inpainting Error Maximization
Figure 3 for Information-Theoretic Segmentation by Inpainting Error Maximization
Figure 4 for Information-Theoretic Segmentation by Inpainting Error Maximization
Viaarxiv icon

Growing Efficient Deep Networks by Structured Continuous Sparsification

Add code
Jul 30, 2020
Figure 1 for Growing Efficient Deep Networks by Structured Continuous Sparsification
Figure 2 for Growing Efficient Deep Networks by Structured Continuous Sparsification
Figure 3 for Growing Efficient Deep Networks by Structured Continuous Sparsification
Figure 4 for Growing Efficient Deep Networks by Structured Continuous Sparsification
Viaarxiv icon

Kernel and Rich Regimes in Overparametrized Models

Add code
Feb 24, 2020
Figure 1 for Kernel and Rich Regimes in Overparametrized Models
Figure 2 for Kernel and Rich Regimes in Overparametrized Models
Figure 3 for Kernel and Rich Regimes in Overparametrized Models
Figure 4 for Kernel and Rich Regimes in Overparametrized Models
Viaarxiv icon

Domain-independent Dominance of Adaptive Methods

Add code
Dec 10, 2019
Figure 1 for Domain-independent Dominance of Adaptive Methods
Figure 2 for Domain-independent Dominance of Adaptive Methods
Figure 3 for Domain-independent Dominance of Adaptive Methods
Figure 4 for Domain-independent Dominance of Adaptive Methods
Viaarxiv icon

Winning the Lottery with Continuous Sparsification

Add code
Dec 10, 2019
Figure 1 for Winning the Lottery with Continuous Sparsification
Figure 2 for Winning the Lottery with Continuous Sparsification
Figure 3 for Winning the Lottery with Continuous Sparsification
Figure 4 for Winning the Lottery with Continuous Sparsification
Viaarxiv icon

Building a Massive Corpus for Named Entity Recognition using Free Open Data Sources

Add code
Aug 13, 2019
Figure 1 for Building a Massive Corpus for Named Entity Recognition using Free Open Data Sources
Figure 2 for Building a Massive Corpus for Named Entity Recognition using Free Open Data Sources
Figure 3 for Building a Massive Corpus for Named Entity Recognition using Free Open Data Sources
Figure 4 for Building a Massive Corpus for Named Entity Recognition using Free Open Data Sources
Viaarxiv icon

On the Convergence of AdaBound and its Connection to SGD

Add code
Aug 13, 2019
Figure 1 for On the Convergence of AdaBound and its Connection to SGD
Viaarxiv icon