The search for effective and robust metrics has been the focus of recent theoretical and empirical work on generalization of deep neural networks (NNs). In this paper, we discuss the performance of natural language processing (NLP) models, and we evaluate various existing and novel generalization metrics. Compared to prior studies, we (i) focus on NLP instead of computer vision (CV), (ii) focus on generalization metrics that predict test error instead of the generalization gap, (iii) focus on generalization metrics that do not need the access to data, and (iv) focus on the heavy-tail (HT) phenomenon that has received comparatively less attention in the study of NNs. We extend recent HT-based work which focuses on power law (PL) distribution...
As deep learning has become solution for various machine learning, artificial intelligence applicati...
The generalization mystery in deep learning is the following: Why do over-parameterized neural netwo...
In recent years Deep Neural Networks (DNNs) have achieved state-of-the-art results in many fields su...
Neural network models have been very successful in natural language inference, with the best models ...
With a direct analysis of neural networks, this paper presents a mathematically tight generalization...
Although recent works have brought some insights into the performance improvement of techniques used...
Modern deep neural networks (DNNs) represent a formidable challenge for theorists: according to the ...
This thesis is concerned with the topic of generalization in large, over-parameterized machine learn...
This file contains all the generalization metrics that can be used to reproduce the results of "Eval...
Estimating the Generalization Error (GE) of Deep Neural Networks (DNNs) is an important task that of...
Deep networks are usually trained and tested in a regime in which the training classification error ...
A machine learning (ML) system must learn not only to match the output of a target function on a tra...
Deep learning has transformed computer vision, natural language processing, and speech recognition. ...
While neural network-based models have achieved impressive performance on a large body of NLP tasks,...
Performance and generalization ability are two important aspects to evaluate deep learning models. H...
As deep learning has become solution for various machine learning, artificial intelligence applicati...
The generalization mystery in deep learning is the following: Why do over-parameterized neural netwo...
In recent years Deep Neural Networks (DNNs) have achieved state-of-the-art results in many fields su...
Neural network models have been very successful in natural language inference, with the best models ...
With a direct analysis of neural networks, this paper presents a mathematically tight generalization...
Although recent works have brought some insights into the performance improvement of techniques used...
Modern deep neural networks (DNNs) represent a formidable challenge for theorists: according to the ...
This thesis is concerned with the topic of generalization in large, over-parameterized machine learn...
This file contains all the generalization metrics that can be used to reproduce the results of "Eval...
Estimating the Generalization Error (GE) of Deep Neural Networks (DNNs) is an important task that of...
Deep networks are usually trained and tested in a regime in which the training classification error ...
A machine learning (ML) system must learn not only to match the output of a target function on a tra...
Deep learning has transformed computer vision, natural language processing, and speech recognition. ...
While neural network-based models have achieved impressive performance on a large body of NLP tasks,...
Performance and generalization ability are two important aspects to evaluate deep learning models. H...
As deep learning has become solution for various machine learning, artificial intelligence applicati...
The generalization mystery in deep learning is the following: Why do over-parameterized neural netwo...
In recent years Deep Neural Networks (DNNs) have achieved state-of-the-art results in many fields su...