Alert button
Picture for Visvanathan Ramesh

Visvanathan Ramesh

Alert button

Designing a Hybrid Neural System to Learn Real-world Crack Segmentation from Fractal-based Simulation

Sep 18, 2023
Achref Jaziri, Martin Mundt, Andres Fernandez Rodriguez, Visvanathan Ramesh

Figure 1 for Designing a Hybrid Neural System to Learn Real-world Crack Segmentation from Fractal-based Simulation
Figure 2 for Designing a Hybrid Neural System to Learn Real-world Crack Segmentation from Fractal-based Simulation
Figure 3 for Designing a Hybrid Neural System to Learn Real-world Crack Segmentation from Fractal-based Simulation
Figure 4 for Designing a Hybrid Neural System to Learn Real-world Crack Segmentation from Fractal-based Simulation
Viaarxiv icon

A Procedural World Generation Framework for Systematic Evaluation of Continual Learning

Jun 04, 2021
Timm Hess, Martin Mundt, Iuliia Pliushch, Visvanathan Ramesh

Figure 1 for A Procedural World Generation Framework for Systematic Evaluation of Continual Learning
Figure 2 for A Procedural World Generation Framework for Systematic Evaluation of Continual Learning
Figure 3 for A Procedural World Generation Framework for Systematic Evaluation of Continual Learning
Figure 4 for A Procedural World Generation Framework for Systematic Evaluation of Continual Learning
Viaarxiv icon

When Deep Classifiers Agree: Analyzing Correlations between Learning Order and Image Statistics

May 19, 2021
Iuliia Pliushch, Martin Mundt, Nicolas Lupp, Visvanathan Ramesh

Figure 1 for When Deep Classifiers Agree: Analyzing Correlations between Learning Order and Image Statistics
Figure 2 for When Deep Classifiers Agree: Analyzing Correlations between Learning Order and Image Statistics
Figure 3 for When Deep Classifiers Agree: Analyzing Correlations between Learning Order and Image Statistics
Figure 4 for When Deep Classifiers Agree: Analyzing Correlations between Learning Order and Image Statistics
Viaarxiv icon

Neural Architecture Search of Deep Priors: Towards Continual Learning without Catastrophic Interference

Apr 14, 2021
Martin Mundt, Iuliia Pliushch, Visvanathan Ramesh

Figure 1 for Neural Architecture Search of Deep Priors: Towards Continual Learning without Catastrophic Interference
Figure 2 for Neural Architecture Search of Deep Priors: Towards Continual Learning without Catastrophic Interference
Figure 3 for Neural Architecture Search of Deep Priors: Towards Continual Learning without Catastrophic Interference
Figure 4 for Neural Architecture Search of Deep Priors: Towards Continual Learning without Catastrophic Interference
Viaarxiv icon

A Wholistic View of Continual Learning with Deep Neural Networks: Forgotten Lessons and the Bridge to Active and Open World Learning

Sep 11, 2020
Martin Mundt, Yong Won Hong, Iuliia Pliushch, Visvanathan Ramesh

Figure 1 for A Wholistic View of Continual Learning with Deep Neural Networks: Forgotten Lessons and the Bridge to Active and Open World Learning
Figure 2 for A Wholistic View of Continual Learning with Deep Neural Networks: Forgotten Lessons and the Bridge to Active and Open World Learning
Figure 3 for A Wholistic View of Continual Learning with Deep Neural Networks: Forgotten Lessons and the Bridge to Active and Open World Learning
Figure 4 for A Wholistic View of Continual Learning with Deep Neural Networks: Forgotten Lessons and the Bridge to Active and Open World Learning
Viaarxiv icon

Fundamental Issues Regarding Uncertainties in Artificial Neural Networks

Feb 25, 2020
Neil A. Thacker, Carole J. Twining, Paul D. Tar, Scott Notley, Visvanathan Ramesh

Figure 1 for Fundamental Issues Regarding Uncertainties in Artificial Neural Networks
Figure 2 for Fundamental Issues Regarding Uncertainties in Artificial Neural Networks
Figure 3 for Fundamental Issues Regarding Uncertainties in Artificial Neural Networks
Figure 4 for Fundamental Issues Regarding Uncertainties in Artificial Neural Networks
Viaarxiv icon

Open Set Recognition Through Deep Neural Network Uncertainty: Does Out-of-Distribution Detection Require Generative Classifiers?

Aug 26, 2019
Martin Mundt, Iuliia Pliushch, Sagnik Majumder, Visvanathan Ramesh

Figure 1 for Open Set Recognition Through Deep Neural Network Uncertainty: Does Out-of-Distribution Detection Require Generative Classifiers?
Figure 2 for Open Set Recognition Through Deep Neural Network Uncertainty: Does Out-of-Distribution Detection Require Generative Classifiers?
Figure 3 for Open Set Recognition Through Deep Neural Network Uncertainty: Does Out-of-Distribution Detection Require Generative Classifiers?
Figure 4 for Open Set Recognition Through Deep Neural Network Uncertainty: Does Out-of-Distribution Detection Require Generative Classifiers?
Viaarxiv icon

Unified Probabilistic Deep Continual Learning through Generative Replay and Open Set Recognition

May 28, 2019
Martin Mundt, Sagnik Majumder, Iuliia Pliushch, Visvanathan Ramesh

Figure 1 for Unified Probabilistic Deep Continual Learning through Generative Replay and Open Set Recognition
Figure 2 for Unified Probabilistic Deep Continual Learning through Generative Replay and Open Set Recognition
Figure 3 for Unified Probabilistic Deep Continual Learning through Generative Replay and Open Set Recognition
Figure 4 for Unified Probabilistic Deep Continual Learning through Generative Replay and Open Set Recognition
Viaarxiv icon

Meta-learning Convolutional Neural Architectures for Multi-target Concrete Defect Classification with the COncrete DEfect BRidge IMage Dataset

Apr 02, 2019
Martin Mundt, Sagnik Majumder, Sreenivas Murali, Panagiotis Panetsos, Visvanathan Ramesh

Figure 1 for Meta-learning Convolutional Neural Architectures for Multi-target Concrete Defect Classification with the COncrete DEfect BRidge IMage Dataset
Figure 2 for Meta-learning Convolutional Neural Architectures for Multi-target Concrete Defect Classification with the COncrete DEfect BRidge IMage Dataset
Figure 3 for Meta-learning Convolutional Neural Architectures for Multi-target Concrete Defect Classification with the COncrete DEfect BRidge IMage Dataset
Figure 4 for Meta-learning Convolutional Neural Architectures for Multi-target Concrete Defect Classification with the COncrete DEfect BRidge IMage Dataset
Viaarxiv icon