{"url":"/method/mixup","slug":"mixup","name":"Mixup","full_name":"Mixup","full_name_withheld":false,"description_markdown":"**Mixup** is a data augmentation technique that generates a weighted combination of random image pairs from the training data. Given two images and their ground truth labels: $\\left(x\\_{i}, y\\_{i}\\right), \\left(x\\_{j}, y\\_{j}\\right)$, a synthetic training example $\\left(\\hat{x}, \\hat{y}\\right)$ is generated as:\r\n\r\n$$ \\hat{x} = \\lambda{x\\_{i}} + \\left(1 − \\lambda\\right){x\\_{j}} $$\r\n$$ \\hat{y} = \\lambda{y\\_{i}} + \\left(1 − \\lambda\\right){y\\_{j}} $$\r\n\r\nwhere $\\lambda \\sim \\text{Beta}\\left(\\alpha = 0.2\\right)$ is independently sampled for each augmented example.","description_state":"present","introduced_year":null,"introduced_by":{"title":"mixup: Beyond Empirical Risk Minimization","paper":"/paper/mixup-beyond-empirical-risk-minimization","first_author":"Hongyi Zhang","n_authors":4,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/mixup-beyond-empirical-risk-minimization"},"source":{"url":"http://arxiv.org/abs/1710.09412v2","title":"mixup: Beyond Empirical Risk Minimization","url_on_a_paper_host":true},"code_snippet_url":"https://github.com/facebookresearch/mixup-cifar10","code_snippet_url_on_a_code_host":true,"categories":[{"area":"Computer Vision","area_id":"computer-vision","collection":"Image Data Augmentation","url":"/methods/category/image-data-augmentation","pwc_aliases":[]}],"n_papers_tagged":651,"archive_num_papers":651,"papers_newest_first":[{"paper":null,"title":"Continual Self-Supervised Learning with Masked Autoencoders in Remote Sensing","date":"2025-06-26","arxiv_id":"2506.21312","n_code_links":0,"syntology":null},{"paper":null,"title":"Enhancing Ambiguous Dynamic Facial Expression Recognition with Soft Label-based Data Augmentation","date":"2025-06-25","arxiv_id":"2506.20867","n_code_links":0,"syntology":null},{"paper":null,"title":"Quantum-Informed Contrastive Learning with Dynamic Mixup Augmentation for Class-Imbalanced Expert Systems","date":"2025-06-16","arxiv_id":"2506.13987","n_code_links":0,"syntology":null},{"paper":"/paper/medusa-a-multimodal-deep-fusion-multi-stage","title":"MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions","date":"2025-06-11","arxiv_id":"2506.09556","n_code_links":1,"syntology":null},{"paper":"/paper/unleashing-the-power-of-intermediate-domains","title":"Unleashing the Power of Intermediate Domains for Mixed Domain Semi-Supervised Medical Image Segmentation","date":"2025-05-30","arxiv_id":"2505.24567","n_code_links":1,"syntology":null},{"paper":"/paper/blast-balanced-sampling-time-series-corpus","title":"BLAST: Balanced Sampling Time Series Corpus for Universal Forecasting Models","date":"2025-05-23","arxiv_id":"2505.17871","n_code_links":2,"syntology":{"ran":5,"of":5,"unverified":0,"pointer_only":0}},{"paper":"/paper/temporal-consistency-constrained-transferable","title":"Temporal Consistency Constrained Transferable Adversarial Attacks with Background Mixup for Action Recognition","date":"2025-05-23","arxiv_id":"2505.17807","n_code_links":1,"syntology":null},{"paper":null,"title":"Cross-Domain Diffusion with Progressive Alignment for Efficient Adaptive Retrieval","date":"2025-05-20","arxiv_id":"2505.13907","n_code_links":0,"syntology":null},{"paper":null,"title":"Suicide Risk Assessment Using Multimodal Speech Features: A Study on the SW1 Challenge Dataset","date":"2025-05-19","arxiv_id":"2505.13069","n_code_links":0,"syntology":null},{"paper":"/paper/saint-attention-based-modeling-of-sub-action","title":"SAINT: Attention-Based Modeling of Sub-Action Dependencies in Multi-Action Policies","date":"2025-05-17","arxiv_id":"2505.12109","n_code_links":1,"syntology":null},{"paper":"/paper/data-agnostic-augmentations-for-unknown","title":"Data-Agnostic Augmentations for Unknown Variations: Out-of-Distribution Generalisation in MRI Segmentation","date":"2025-05-15","arxiv_id":"2505.10223","n_code_links":1,"syntology":null},{"paper":"/paper/uwav-uncertainty-weighted-weakly-supervised","title":"UWAV: Uncertainty-weighted Weakly-supervised Audio-Visual Video Parsing","date":"2025-05-14","arxiv_id":"2505.09615","n_code_links":1,"syntology":null},{"paper":null,"title":"GradMix: Gradient-based Selective Mixup for Robust Data Augmentation in Class-Incremental Learning","date":"2025-05-13","arxiv_id":"2505.08528","n_code_links":0,"syntology":null},{"paper":"/paper/mediaug-exploring-visual-augmentation-in","title":"MediAug: Exploring Visual Augmentation in Medical Imaging","date":"2025-04-26","arxiv_id":"2504.18983","n_code_links":1,"syntology":null},{"paper":null,"title":"AMAD: AutoMasked Attention for Unsupervised Multivariate Time Series Anomaly Detection","date":"2025-04-09","arxiv_id":"2504.06643","n_code_links":0,"syntology":null},{"paper":"/paper/mixeeg-enhancing-eeg-federated-learning-for","title":"mixEEG: Enhancing EEG Federated Learning for Cross-subject EEG Classification with Tailored mixup","date":"2025-04-07","arxiv_id":"2504.07987","n_code_links":1,"syntology":null},{"paper":null,"title":"iADCPS: Time Series Anomaly Detection for Evolving Cyber-physical Systems via Incremental Meta-learning","date":"2025-04-06","arxiv_id":"2504.04374","n_code_links":0,"syntology":null},{"paper":null,"title":"Enlightenment Period Improving DNN Performance","date":"2025-04-02","arxiv_id":"2504.01737","n_code_links":0,"syntology":null},{"paper":null,"title":"ElimPCL: Eliminating Noise Accumulation with Progressive Curriculum Labeling for Source-Free Domain Adaptation","date":"2025-03-31","arxiv_id":"2503.23712","n_code_links":0,"syntology":null},{"paper":"/paper/recurrent-feature-mining-and-keypoint-mixup","title":"Recurrent Feature Mining and Keypoint Mixup Padding for Category-Agnostic Pose Estimation","date":"2025-03-27","arxiv_id":"2503.21140","n_code_links":1,"syntology":null},{"paper":null,"title":"V2X-DG: Domain Generalization for Vehicle-to-Everything Cooperative Perception","date":"2025-03-19","arxiv_id":"2503.15435","n_code_links":0,"syntology":null},{"paper":"/paper/similarity-aware-token-pruning-your-vlm-but","title":"Similarity-Aware Token Pruning: Your VLM but Faster","date":"2025-03-14","arxiv_id":"2503.11549","n_code_links":1,"syntology":null},{"paper":null,"title":"Class-Aware PillarMix: Can Mixed Sample Data Augmentation Enhance 3D Object Detection with Radar Point Clouds?","date":"2025-03-04","arxiv_id":"2503.02687","n_code_links":0,"syntology":null},{"paper":null,"title":"Semi-Supervised Audio-Visual Video Action Recognition with Audio Source Localization Guided Mixup","date":"2025-03-04","arxiv_id":"2503.02284","n_code_links":0,"syntology":null},{"paper":"/paper/a-generalized-theory-of-mixup-for-structure","title":"A Generalized Theory of Mixup for Structure-Preserving Synthetic Data","date":"2025-03-03","arxiv_id":"2503.02645","n_code_links":1,"syntology":null},{"paper":null,"title":"MIDAS: Mixing Ambiguous Data with Soft Labels for Dynamic Facial Expression Recognition","date":"2025-02-28","arxiv_id":"2503.00202","n_code_links":0,"syntology":null},{"paper":null,"title":"Single Domain Generalization with Model-aware Parametric Batch-wise Mixup","date":"2025-02-22","arxiv_id":"2502.16064","n_code_links":0,"syntology":null},{"paper":"/paper/mixup-model-merge-enhancing-model-merging","title":"Mixup Model Merge: Enhancing Model Merging Performance through Randomized Linear Interpolation","date":"2025-02-21","arxiv_id":"2502.15434","n_code_links":1,"syntology":null},{"paper":null,"title":"TimeDistill: Efficient Long-Term Time Series Forecasting with MLP via Cross-Architecture Distillation","date":"2025-02-20","arxiv_id":"2502.15016","n_code_links":0,"syntology":null},{"paper":"/paper/dual-level-mixup-for-graph-few-shot-learning","title":"Dual-level Mixup for Graph Few-shot Learning with Fewer Tasks","date":"2025-02-19","arxiv_id":"2502.14158","n_code_links":1,"syntology":null}],"papers_shown":30,"tasks":[{"task":"/task/data-augmentation","name":"Data Augmentation","papers":245},{"task":"/task/image-classification","name":"Image Classification","papers":84},{"task":"/task/domain-adaptation","name":"Domain Adaptation","papers":66},{"task":"/task/image-classification","name":"image-classification","papers":63},{"task":"/task/classification-1","name":"Classification","papers":40},{"task":"/task/contrastive-learning","name":"Contrastive Learning","papers":38},{"task":"/task/representation-learning","name":"Representation Learning","papers":35},{"task":"/task/unsupervised-domain-adaptation","name":"Unsupervised Domain Adaptation","papers":33},{"task":"/task/transfer-learning","name":"Transfer Learning","papers":32},{"task":"/task/diversity","name":"Diversity","papers":28},{"task":"/task/object-detection","name":"Object Detection","papers":27},{"task":"/task/semantic-segmentation","name":"Semantic Segmentation","papers":27},{"task":"/task/classification","name":"General Classification","papers":25},{"task":"/task/knowledge-distillation","name":"Knowledge Distillation","papers":25},{"task":"/task/domain-generalization","name":"Domain Generalization","papers":23},{"task":"/task/object-detection-1","name":"object-detection","papers":22},{"task":"/task/self-supervised-learning","name":"Self-Supervised Learning","papers":21},{"task":"/task/segmentation","name":"Segmentation","papers":19},{"task":"/task/text-classification","name":"Text Classification","papers":17},{"task":"/task/graph-classification","name":"Graph Classification","papers":16}],"tasks_shown":20,"n_tasks":424,"usage_by_year":[{"year":"2017","papers":1},{"year":"2018","papers":7},{"year":"2019","papers":32},{"year":"2020","papers":67},{"year":"2021","papers":121},{"year":"2022","papers":134},{"year":"2023","papers":137},{"year":"2024","papers":106},{"year":"2025","papers":46}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/mixup"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}