Limited Sample

Limited sample learning addresses the challenge of building accurate models with insufficient training data, a common problem across diverse fields. Current research focuses on developing robust algorithms and model architectures, such as contrastive learning, self-training, and meta-learning, to improve performance in these scenarios, often incorporating techniques like uncertainty quantification and data augmentation to maximize the information extracted from limited samples. This research is crucial for advancing applications where obtaining large datasets is expensive or impossible, including biomedical image analysis, emotion recognition, and reinforcement learning in real-world settings. The development of effective limited sample learning methods has significant implications for improving the reliability and applicability of machine learning across various scientific disciplines and practical applications.

Papers

September 12, 2024

From Uncertainty to Clarity: Uncertainty-Guided Class-Incremental Learning for Limited Biomedical Samples via Semantic Expansion
Yifei Yao, Hanrong Zhang
Strong Generalization High Uncertainty Anticipation Incremental Learning Class Incremental Learning Entropy Model Limited Sample

August 23, 2024

Leveraging Contrastive Learning and Self-Training for Multimodal Emotion Recognition with Limited Labeled Samples
Qi Fan, Yutong Li, Yi Xin, Xinyu Cheng, Guanglai Gao, Miao Ma
Contrastive Learning Semi Supervised Learning Emotion Recognition Self Training Multimodal Emotion Recognition Modality Representation Limited Sample

August 22, 2024

Domain Adaptation for Offline Reinforcement Learning with Limited Samples
Weiqin Chen, Sandipan Mishra, Santiago Paternain
Domain Adaptation Offline Reinforcement Learning Offline Reinforcement Learning Algorithm Limited Sample Target Dataset

January 6, 2024

Human-Instruction-Free LLM Self-Alignment with Limited Samples
Hongyi Guo, Yuanshun Yao, Wei Shen, Jiaheng Wei, Xiaoying Zhang, Zhaoran Wang, Yang Liu
Large Language Model Domain Adaptive LLM Alignment Limited Sample

December 22, 2023

HyperMix: Out-of-Distribution Detection and Classification in Few-Shot Settings
Nikhil Mehta, Kevin J Liang, Jing Huang, Fu-Jen Chu, Li Yin, Tal Hassner
Classification Code Distribution Detection Distribution Data Outlier Exposure Limited Sample Auxiliary Outlier

November 26, 2023

The Local Landscape of Phase Retrieval Under Limited Samples
Kaizhao Liu, Zihao Wang, Lei Wu
Non Convex Phase Retrieval Global Minimum Limited Sample

November 10, 2023

Learning Contrastive Self-Distillation for Ultra-Fine-Grained Visual Categorization Targeting Limited Samples
Ziye Fang, Xin Jiang, Hao Tang, Zechao Li
Contrastive Learning Contrastive Distillation Discrepancy Aware Limited Sample

October 9, 2023

Estimating Shape Distances on Neural Representations with Limited Samples
Dean A. Pospisil, Brett W. Larsen, Sarah E. Harvey, Alex H. Williams
Neural Representation Shape Estimation Network Representation Limited Sample Shape Metric Geometric Similarity

September 5, 2023

Learning Cross-Modal Affinity for Referring Video Object Segmentation Targeting Limited Samples
Guanghui Li, Mingqi Gao, Heng Liu, Xiantong Zhen, Feng Zheng
LeArning Abstract Referring Video Object Segmentation Limited Sample

August 1, 2023

Multiscale Feature Learning Using Co-Tuplet Loss for Offline Handwritten Signature Verification
Fu-Hsien Huang, Hsin-Min Lu
Signature Verification Limited Sample Multiscale Feature Offline Handwritten Signature Verification Octuplet Loss

March 22, 2023

Strategy Synthesis in Markov Decision Processes Under Limited Sampling Access
Christel Baier, Clemens Dubslaff, Patrick Wienhöft, Stefan J. Kiebel
Reinforcement Learning Markov Decision Process Limited Sample Strategy Synthesis Reward Maximizing Interval Markov Decision Process

February 8, 2023

Enhancing Modality-Agnostic Representations via Meta-Learning for Brain Tumor Segmentation
Aishik Konwer, Xiaoling Hu, Joseph Bae, Xuan Xu, Chao Chen, Prateek Prasanna
Brain Tumor Segmentation Modality Data Modality Representation Limited Sample Modality Agnostic

October 3, 2022

Learning with Limited Samples -- Meta-Learning and Applications to Communication Systems
Lisha Chen, Sharu Theresa Jose, Ivana Nikoloska, Sangwoo Park, Tianyi Chen, Osvaldo Simeone
Deep Learning LeArning Abstract Deep Learning Model Financial Application Meta Learning Joint Learning Communication System Limited Sample Shot Meta Learning

March 25, 2022

On efficient algorithms for computing near-best polynomial approximations to high-dimensional, Hilbert-valued functions from limited samples
Ben Adcock, Simone Brugiapaglia, Nick Dexter, Sebastian Moraga
Compressed Sensing Efficient Algorithm Approximation Rate Polynomial Approximation Limited Sample Sparse Function Hilbert Transform \Ell_1$ Minimization

Limited Sample

Papers

From Uncertainty to Clarity: Uncertainty-Guided Class-Incremental Learning for Limited Biomedical Samples via Semantic Expansion

Leveraging Contrastive Learning and Self-Training for Multimodal Emotion Recognition with Limited Labeled Samples

Domain Adaptation for Offline Reinforcement Learning with Limited Samples

Human-Instruction-Free LLM Self-Alignment with Limited Samples

HyperMix: Out-of-Distribution Detection and Classification in Few-Shot Settings

The Local Landscape of Phase Retrieval Under Limited Samples

Learning Contrastive Self-Distillation for Ultra-Fine-Grained Visual Categorization Targeting Limited Samples

Estimating Shape Distances on Neural Representations with Limited Samples

Learning Cross-Modal Affinity for Referring Video Object Segmentation Targeting Limited Samples

Multiscale Feature Learning Using Co-Tuplet Loss for Offline Handwritten Signature Verification

Strategy Synthesis in Markov Decision Processes Under Limited Sampling Access

Enhancing Modality-Agnostic Representations via Meta-Learning for Brain Tumor Segmentation

Learning with Limited Samples -- Meta-Learning and Applications to Communication Systems

On efficient algorithms for computing near-best polynomial approximations to high-dimensional, Hilbert-valued functions from limited samples