Synthetic Feature

Synthetic features are artificially generated data points designed to augment existing datasets, addressing limitations like data scarcity, class imbalance, and the need for improved model generalization. Current research focuses on generating synthetic features using various methods, including generative adversarial networks (GANs), transformers, and optimal transport, often within specific model architectures like deep feedforward networks and transformers to improve performance in tasks such as image classification, object detection, and fault diagnosis. This work is significant because it enhances the robustness and accuracy of machine learning models, particularly in domains with limited real-world data, leading to improved performance in applications ranging from medical image analysis to industrial fault detection.

Papers

February 4, 2024

Arithmetic Feature Interaction Is Necessary for Deep Tabular Learning
Yi Cheng, Renjun Hu, Haochao Ying, Xing Shi, Jian Wu, Wei Lin
Feature Interaction Tabular Deep Learning Synthetic Feature

December 27, 2023

ReSynthDetect: A Fundus Anomaly Detection Network with Reconstruction and Synthetic Features
Jingqi Niu, Qinji Yu, Shiwen Dong, Zilong Wang, Kang Dang, Xiaowei Ding
Full State Reconstruction Fundus Image Pixel Level Anomaly Synthetic Feature

August 29, 2023

Few-Shot Object Detection via Synthetic Features with Optimal Transport
Anh-Khoa Nguyen Vu, Thanh-Toan Do, Vinh-Tiep Nguyen, Tam Le, Minh-Triet Tran, Tam V. Nguyen
Training Data Synthetic Data Optimal Transport Benchmark Datasets Shot Object Detection Synthetic Feature

July 14, 2023

Improving Zero-Shot Generalization for CLIP with Synthesized Prompts
Zhengbo Wang, Jian Liang, Ran He, Nan Xu, Zilei Wang, Tieniu Tan
Vision Language Model Strong Generalization Variational Autoencoders Single CLIP Generative Approach Generalized Zero Shot Learning Diverse Prompt Synthetic Feature

May 17, 2023

Utility Theory of Synthetic Data Generation
Shirong Xu, Will Wei Sun, Guang Cheng
Synthetic Data Synthetic Data Generation Generalization Gap Utility Based Synthetic Feature Synthetic Model

April 6, 2023

Synthetic Sample Selection for Generalized Zero-Shot Learning
Shreyank N Gowda
Reinforcement Learning Generalized Zero Shot Learning Bidirectional Transformer Sample Generation Synthetic Feature

November 2, 2022

Data-driven design of fault diagnosis for three-phase PWM rectifier using random forests technique with transient synthetic features
Lei Kou, Chuang Liu, Guo-wei Cai, Jia-ning Zhou, Quan-de Yuan
Random Forest Data Driven Design Data Driven Fault Synthetic Feature Open Circuit Fault

November 1, 2022

Fault diagnosis for three-phase PWM rectifier based on deep feedforward network with transient synthetic features
Kou Lei, Liu Chuang, Cai Guo-Wei, Zhang Zhe, Zhou Jia-Ning, Wang Xue-Mei
Neural Network Fault Diagnosis Method Synthetic Feature Transient Fault

September 28, 2022

Data Augmentation using Feature Generation for Volumetric Medical Images
Khushboo Mehra, Hassan Soliman, Soumya Ranjan Sahoo
Deep Learning Model Data Augmentation Medical Imaging Medical Image Classification Feature Generation Volumetric Medical Auxiliary Classifier Synthetic Feature

April 4, 2022

Interpretable Saliency Maps And Self-Supervised Learning For Generalized Zero Shot Medical Image Classification
Dwarikanath Mahapatra
Self Supervised Learning Medical Image Saliency Map Medical Image Classification Natural Image Unseen Class Synthetic Feature

February 8, 2022

Uncertainty Modeling for Out-of-Distribution Generalization
Xiaotong Li, Yongxing Dai, Yixiao Ge, Jun Liu, Ying Shan, Ling-Yu Duan
Domain Shift Distribution Generalization Uncertainty Modeling Feature Uncertainty Statistical Feature Synthetic Feature