Learned Feature

Learned features, extracted from data using machine learning models, are transforming various fields by enabling automated feature extraction surpassing hand-crafted alternatives. Current research focuses on improving the quality and interpretability of these features, exploring architectures like convolutional neural networks (CNNs), transformers, and recurrent neural networks (RNNs), and employing techniques such as self-supervised learning and gradient-based methods. This work is significant because it enhances the performance and reliability of machine learning models across diverse applications, from image analysis and gesture recognition to medical diagnosis and anomaly detection, ultimately leading to more accurate and robust systems.

Papers

October 24, 2024

A Random Matrix Theory Perspective on the Spectrum of Learned Features and Asymptotic Generalization Capabilities
Yatin Dandi, Luca Pesce, Hugo Cui, Florent Krzakala, Yue M. Lu, Bruno Loureiro
Learning Rate Two Layer Neural Network Random Feature Random Matrix Time to Spectrum Fully Connected Neural Network Learned Feature

September 4, 2024

Deep Learning Meets Satellite Images -- An Evaluation on Handcrafted and Learning-based Features for Multi-date Satellite Stereo Images
Shuang Song, Luca Morelli, Xinyi Wu, Rongjun Qin, Hessah Albanwan, Fabio Remondino
Deep Learning Satellite Image Satellite Imagery Feature Matching Traditional CrAFT Multi Temporal Remote Sensing Image Learned Feature Deep Matching

June 18, 2024

Deep self-supervised learning with visualisation for automatic gesture recognition
Fabien Allemand, Alessio Mazzela, Jun Villette, Decky Aspandi, Titus Zaharia
Self Supervised Learning Gesture Recognition Self Supervised Method High Quality Gesture Body Gesture Learned Feature

April 26, 2024

Conformal Prediction with Learned Features
Shayan Kiyani, George Pappas, Hamed Hassani
Conformal Prediction Provable Guarantee Feature Uncertainty Learned Feature

October 19, 2023

Provable Guarantees for Neural Networks via Gradient Feature Learning
Zhenmei Shi, Junyi Wei, Yingyu Liang
Neural Network Gradient Descent Gradient Based Neural Tangent Kernel Feature Learning Provable Guarantee Learned Feature

August 29, 2023

Analysis of Learned Features and Framework for Potato Disease Detection
Shikha Gupta, Soma Chakraborty, Renu Rameshan
Convolutional Neural Network Training Data General Analysis New Framework Plant Disease Attention Based Network Learned Feature Potato Disease

July 15, 2023

Single and Multi-Speaker Cloned Voice Detection: From Perceptual to Learned Features
Sarah Barrington, Romit Barua, Gautham Koorma, Hany Farid
Spectral Feature Synthetic Voice Voice Cloning Perceptual Understanding Learned Feature

June 19, 2023

ExpPoint-MAE: Better interpretability and performance for self-supervised point cloud transformers
Ioannis Romanelis, Vlassis Fotis, Konstantinos Moustakas, Adrian Munteanu
Transformer Based System Performance Transformer Model Learned Representation Better Interpretability Point Cloud Transformer Learned Feature

May 30, 2023

Bottleneck Structure in Learned Features: Low-Dimension vs Regularity Tradeoff
Arthur Jacot
Low Dimensional Representation Large Learning Rate Infinite Depth L2 Regularization Learned Feature Low Dimension Bottleneck Structure

May 17, 2023

A hybrid feature learning approach based on convolutional kernels for ATM fault prediction using event-log data
Víctor Manuel Vargas, Riccardo Rosati, César Hervás-Martínez, Adriano Mancini, Luca Romeo, Pedro Antonio Gutiérrez
Constructive Approach Time Series Classification Convolutional Kernel Learned Feature Hybrid Feature Event Log Fault Prediction

November 16, 2022

Interpretable Few-shot Learning with Online Attribute Selection
Mohammad Reza Zarei, Majid Komeili
Shot Learning Interpretable Model Shot Classification Learned Feature Attribute Selection

October 20, 2022

Freeze then Train: Towards Provable Representation Learning under Spurious Correlations and Feature Noise
Haotian Ye, James Zou, Linjun Zhang
Empirical Risk Minimization Spurious Correlation High Speed Train System Learned Feature Adaptive Freezing Feature Noise

June 16, 2022

Open-Set Recognition with Gradient-Based Representations
Jinsol Lee, Ghassan AlRegib
Open Set Open Set Recognition Image Classification Task Gradient Alignment Learned Feature

May 10, 2022

Using Deep Learning-based Features Extracted from CT scans to Predict Outcomes in COVID-19 Patients
Sai Vidyaranya Nuthalapati, Marcela Vizcaychipi, Pallav Shah, Piotr Chudzik, Chee Hau Leow, Paria Yousefi, Ahmed Selim, Keiran Tait, Ben Irving
Computed Tomography Multiple Outcome Intensive Care Unit Covid 19 Patient Mortality Prediction Learned Feature Qualitative Feature Intensive Care Unit Mortality

April 4, 2022

Analyzing the Effects of Handling Data Imbalance on Learned Features from Medical Images by Looking Into the Models
Ashkan Khakzar, Yawei Li, Yang Zhang, Mirac Sanisoglu, Seong Tae Kim, Mina Rezaei, Bernd Bischl, Nassir Navab
Convolutional Neural Network Full Model Medical Image Mixed Effect Data Imbalance Medical Datasets Group Imbalance Imbalanced Dataset Learned Feature

Learned Feature

Papers

A Random Matrix Theory Perspective on the Spectrum of Learned Features and Asymptotic Generalization Capabilities

Deep Learning Meets Satellite Images -- An Evaluation on Handcrafted and Learning-based Features for Multi-date Satellite Stereo Images

Deep self-supervised learning with visualisation for automatic gesture recognition

Conformal Prediction with Learned Features

Provable Guarantees for Neural Networks via Gradient Feature Learning

Analysis of Learned Features and Framework for Potato Disease Detection

Single and Multi-Speaker Cloned Voice Detection: From Perceptual to Learned Features

ExpPoint-MAE: Better interpretability and performance for self-supervised point cloud transformers

Bottleneck Structure in Learned Features: Low-Dimension vs Regularity Tradeoff

A hybrid feature learning approach based on convolutional kernels for ATM fault prediction using event-log data

Interpretable Few-shot Learning with Online Attribute Selection

Freeze then Train: Towards Provable Representation Learning under Spurious Correlations and Feature Noise

Open-Set Recognition with Gradient-Based Representations

Using Deep Learning-based Features Extracted from CT scans to Predict Outcomes in COVID-19 Patients

Analyzing the Effects of Handling Data Imbalance on Learned Features from Medical Images by Looking Into the Models