Awesome！多任务学习宝藏资料
每天给你送来NLP技术干货！
作者 | WeiHongLee
整理 | NewBeeNLP
最近冲浪发现一份多任务学习宝藏资料，包含了领域综述、论文、代码等内容，分享给大家。感谢作者的整理，有帮助记得start噢！
https://github.com/WeiHongLee/Awesome-Multi-Task-Learning
Table of Contents
Survey & Study Benchmarks & Code Papers Awesome Multi-domain Multi-task Learning Workshops Online Courses Related awesome list
Survey & Study
Factors of Influence for Transfer Learning across Diverse Appearance Domains and Task Types (TPAMI, 2022) [paper[1]]
Multi-Task Learning for Dense Prediction Tasks: A Survey (TPAMI, 2021) [paper[2]] [code[3]]
A Survey on Multi-Task Learning (TKDE, 2021) [paper[4]]
Multi-Task Learning with Deep Neural Networks: A Survey (arXiv, 2020) [paper[5]]
Taskonomy: Disentangling Task Transfer Learning (CVPR, 2018 [best paper]) [paper[6]] [dataset[7]]
A Comparison of Loss Weighting Strategies for Multi task Learning in Deep Neural Networks (IEEE Access, 2019) [paper[8]]
An Overview of Multi-Task Learning in Deep Neural Networks (arXiv, 2017) [paper[9]]
Benchmarks & Code
Benchmarks
Dense Prediction Tasks
[NYUv2] Indoor Segmentation and Support Inference from RGBD Images (ECCV, 2012) [paper[10]] [dataset[11]]
[Cityscapes] The Cityscapes Dataset for Semantic Urban Scene Understanding (CVPR, 2016) [paper[12]] [dataset[13]]
[PASCAL-Context] The Role of Context for Object Detection and Semantic Segmentation in the Wild (CVPR, 2014) [paper[14]] [dataset[15]]
[Taskonomy] Taskonomy: Disentangling Task Transfer Learning (CVPR, 2018 [best paper]) [paper[16]] [dataset[17]]
[KITTI] Vision meets robotics: The KITTI dataset (IJRR, 2013) [paper[18]] dataset[19]
[SUN RGB-D] SUN RGB-D: A RGB-D Scene Understanding Benchmark Suite (CVPR 2015) [paper[20]] [dataset[21]]
[BDD100K] BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning (CVPR, 2020) [paper[22]] [dataset[23]]
[Omnidata] Omnidata: A Scalable Pipeline for Making Multi-Task Mid-Level Vision Datasets from 3D Scans (ICCV, 2021) [paper[24]] [project[25]]
Image Classification
[Meta-dataset] Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples (ICLR, 2020) [paper[26]] [dataset[27]]
[Visual Domain Decathlon] Learning multiple visual domains with residual adapters (NeurIPS, 2017) [paper[28]] [dataset[29]]
[CelebA] Deep Learning Face Attributes in the Wild (ICCV, 2015) [paper[30]] [dataset[31]]
Code
[Multi-Task-Learning-PyTorch[32]]: Multi-task Dense Prediction.
[Auto-λ[33]]: Multi-task Dense Prediction, Robotics.
[UniversalRepresentations[34]]: Multi-task Dense Prediction[35] (including different loss weighting strategies), Multi-domain Classification[36], Cross-domain Few-shot Learning[37].
[MTAN[38]]: Multi-task Dense Prediction, Multi-domain Classification.
[ASTMT[39]]: Multi-task Dense Prediction.
[LibMTL[40]]: Multi-task Dense Prediction.
[MTPSL[41]]: Multi-task Partially-supervised Learning for Dense Prediction.
[Resisual Adapater[42]]: Multi-domain Classification.
Papers
2022
Task Discovery: Finding the Tasks that Neural Networks Generalize on (NeurIPS, 2022)
[Auto-λ] Auto-λ: Disentangling Dynamic Task Relationships (TMLR, 2022) [paper[43]] [code[44]]
[Universal Representations] Universal Representations: A Unified Look at Multiple Task and Domain Learning (arXiv, 2022) [paper[45]] [code[46]]
Not All Models Are Equal: Predicting Model Transferability in a Self-challenging Fisher Space (ECCV, 2022) [paper[47]] [code[48]]
Factorizing Knowledge in Neural Networks (ECCV, 2022) [paper[49]] [code[50]]
[InvPT] Inverted Pyramid Multi-task Transformer for Dense Scene Understanding (ECCV, 2022) [paper[51]] [code[52]]
[MultiMAE] MultiMAE: Multi-modal Multi-task Masked Autoencoders (ECCV, 2022) [paper[53]] [code[54]]
A Multi-objective / Multi-task Learning Framework Induced by Pareto Stationarity (ICML, 2022) [paper[55]]
Mitigating Modality Collapse in Multimodal VAEs via Impartial Optimization (ICML, 2022) [paper[56]]
Active Multi-Task Representation Learning (ICML, 2022) [paper[57]]
Generative Modeling for Multi-task Visual Learning (ICML, 2022) [paper[58]] [code[59]]
Multi-Task Learning as a Bargaining Game (ICML, 2022) [paper[60]] [code[61]]
Multi-Task Learning with Multi-query Transformer for Dense Prediction (arXiv, 2022) [paper[62]]
[Gato] A Generalist Agent (arXiv, 2022) [paper[63]]
[MTPSL] Learning Multiple Dense Prediction Tasks from Partially Annotated Data (CVPR, 2022) [paper[64]] [code[65]]
[TSA] Cross-domain Few-shot Learning with Task-specific Adapters (CVPR, 2022) [paper[66]] [code[67]]
[OMNIVORE] OMNIVORE: A Single Model for Many Visual Modalities (CVPR, 2022) [paper[68]] [code[69]]
Task Adaptive Parameter Sharing for Multi-Task Learning (CVPR, 2022) [paper[70]]
Controllable Dynamic Multi-Task Architectures (CVPR, 2022) [paper[71]] [code[72]]
[SHIFT] SHIFT: A Synthetic Driving Dataset for Continuous Multi-Task Domain Adaptation (CVPR, 2022) [paper[73]] [code[74]]
DiSparse: Disentangled Sparsification for Multitask Model Compression (CVPR, 2022) [paper[75]] [code[76]]
[MulT] MulT: An End-to-End Multitask Learning Transformer (CVPR, 2022) [paper[77]] [code[78]]
Sound and Visual Representation Learning with Multiple Pretraining Tasks (CVPR, 2022) [paper[79]]
Medusa: Universal Feature Learning via Attentional Multitasking (CVPR Workshop, 2022) [paper[80]]
An Evolutionary Approach to Dynamic Introduction of Tasks in Large-scale Multitask Learning Systems (arXiv, 2022) [paper[81]] [code[82]]
Combining Modular Skills in Multitask Learning (arXiv, 2022) [paper[83]]
Visual Representation Learning over Latent Domains (ICLR, 2022) [paper[84]]
ADARL: What, Where, and How to Adapt in Transfer Reinforcement Learning (ICLR, 2022) [paper[85]] [code[86]]
Towards a Unified View of Parameter-Efficient Transfer Learning (ICLR, 2022) [paper[87]] [code[88]]
[Rotograd] Rotograd: Dynamic Gradient Homogenization for Multi-Task Learning (ICLR, 2022) [paper[89]] [code[90]]
Relational Multi-task Learning: Modeling Relations Between Data and Tasks (ICLR, 2022) [paper[91]]
Weighted Training for Cross-task Learning (ICLR, 2022) [paper[92]] [code[93]]
Semi-supervised Multi-task Learning for Semantics and Depth (WACV, 2022) [paper[94]]
In Defense of the Unitary Scalarization for Deep Multi-Task Learning (arXiv, 2022) [paper[95]]
2021
Efficiently Identifying Task Groupings for Multi-Task Learning (NeurIPS, 2021) [paper[96]]
[CAGrad] Conflict-Averse Gradient Descent for Multi-task Learning (NeurIPS, 2021) [paper[97]] [code[98]]
A Closer Look at Loss Weighting in Multi-Task Learning (arXiv, 2021) [paper[99]]
Exploring Relational Context for Multi-Task Dense Prediction (ICCV, 2021) [paper[100]] [code[101]]
Multi-Task Self-Training for Learning General Representations (ICCV, 2021) [paper[102]]
Task Switching Network for Multi-task Learning (ICCV, 2021) [paper[103]] [code[104]]
Omnidata: A Scalable Pipeline for Making Multi-Task Mid-Level Vision Datasets from 3D Scans (ICCV, 2021) [paper[105]] [project[106]]
Robustness via Cross-Domain Ensembles (ICCV, 2021) [paper[107]] [code[108]]
Domain Adaptive Semantic Segmentation with Self-Supervised Depth Estimation (ICCV, 2021) [paper[109]] [code[110]]
[URL] Universal Representation Learning from Multiple Domains for Few-shot Classification (ICCV, 2021) [paper[111]] [code[112]]
[tri-M] A Multi-Mode Modulator for Multi-Domain Few-Shot Classification (ICCV, 2021) [paper[113]] [code[114]]
MultiTask-CenterNet (MCN): Efficient and Diverse Multitask Learning using an Anchor Free Approach (ICCV Workshop, 2021) [paper[115]]
See Yourself in Others: Attending Multiple Tasks for Own Failure Detection (arXiv, 2021) [paper[116]]
A Multi-Task Cross-Task Learning Architecture for Ad-hoc Uncertainty Estimation in 3D Cardiac MRI Image Segmentation (CinC, 2021) [paper[117]] [code[118]]
Multi-Task Reinforcement Learning with Context-based Representations (ICML, 2021) [paper[119]]
[FLUTE] Learning a Universal Template for Few-shot Dataset Generalization (ICML, 2021) [paper[120]] [code[121]]
Towards a Unified View of Parameter-Efficient Transfer Learning (arXiv, 2021) [paper[122]]
UniT: Multimodal Multitask Learning with a Unified Transformer (arXiv, 2021) [paper[123]]
Learning to Relate Depth and Semantics for Unsupervised Domain Adaptation (CVPR, 2021) [paper[124]] [code[125]]
CompositeTasking: Understanding Images by Spatial Composition of Tasks (CVPR, 2021) [paper[126]] [code[127]]
Anomaly Detection in Video via Self-Supervised and Multi-Task Learning (CVPR, 2021) [paper[128]]
Taskology: Utilizing Task Relations at Scale (CVPR, 2021) [paper[129]]
Three Ways to Improve Semantic Segmentation with Self-Supervised Depth Estimation (CVPR, 2021) [paper[130]] [code[131]]
Improving Semi-Supervised and Domain-Adaptive Semantic Segmentation with Self-Supervised Depth Estimation (arXiv, 2021) [paper[132]] [code[133]]
Counter-Interference Adapter for Multilingual Machine Translation (Findings of EMNLP, 2021) [paper[134]]
Conditionally Adaptive Multi-Task Learning: Improving Transfer Learning in NLP Using Fewer Parameters & Less Data (ICLR) [paper[135]] [code[136]]
[Gradient Vaccine] Gradient Vaccine: Investigating and Improving Multi-task Optimization in Massively Multilingual Models (ICLR, 2021) [paper[137]]
[IMTL] Towards Impartial Multi-task Learning (ICLR, 2021) [paper[138]]
Deciphering and Optimizing Multi-Task Learning: A Random Matrix Approach (ICLR, 2021) [paper[139]]
[URT] A Universal Representation Transformer Layer for Few-Shot Image Classification (ICLR, 2021) [paper[140]] [code[141]]
Flexible Multi-task Networks by Learning Parameter Allocation (ICLR Workshop, 2021) [paper[142]]
Multi-Loss Weighting with Coefficient of Variations (WACV, 2021) [paper[143]] [code[144]]
2020
Multi-Task Reinforcement Learning with Soft Modularization (NeurIPS, 2020) [paper[145]] [code[146]]
AdaShare: Learning What To Share For Efficient Deep Multi-Task Learning (NeurIPS, 2020) [paper[147]] [code[148]]
[GradDrop] Just Pick a Sign: Optimizing Deep Multitask Models with Gradient Sign Dropout (NeurIPS, 2020) [paper[149]] [code[150]]
[PCGrad] Gradient Surgery for Multi-Task Learning (NeurIPS, 2020) [paper[151]] [tensorflow[152]] [pytorch[153]]
On the Theory of Transfer Learning: The Importance of Task Diversity (NeurIPS, 2020) [paper[154]]
A Study of Residual Adapters for Multi-Domain Neural Machine Translation (WMT, 2020) [paper[155]]
Multi-Task Adversarial Attack (arXiv, 2020) [paper[156]]
Automated Search for Resource-Efficient Branched Multi-Task Networks (BMVC, 2020) [paper[157]] [code[158]]
Branched Multi-Task Networks: Deciding What Layers To Share (BMVC, 2020) [paper[159]]
MTI-Net: Multi-Scale Task Interaction Networks for Multi-Task Learning (ECCV, 2020) [paper[160]] [code[161]]
Reparameterizing Convolutions for Incremental Multi-Task Learning without Task Interference (ECCV, 2020) [paper[162]] [code[163]]
Selecting Relevant Features from a Multi-domain Representation for Few-shot Classification (ECCV, 2020) [paper[164]] [code[165]]
Multitask Learning Strengthens Adversarial Robustness (ECCV 2020) [paper[166]] [code[167]]
Duality Diagram Similarity: a generic framework for initialization selection in task transfer learning (ECCV, 2020) [paper[168]] [code[169]]
[KD4MTL] Knowledge Distillation for Multi-task Learning (ECCV Workshop) [paper[170]] [code[171]]
MTL-NAS: Task-Agnostic Neural Architecture Search towards General-Purpose Multi-Task Learning (CVPR, 2020) [paper[172]] [code[173]]
Robust Learning Through Cross-Task Consistency (CVPR, 2020) [paper[174]] [code[175]]
12-in-1: Multi-Task Vision and Language Representation Learning (CVPR, 2020) paper[176] [code[177]]
A Multi-task Mean Teacher for Semi-supervised Shadow Detection (CVPR, 2020) [paper[178]] [code[179]]
MAD-X: An Adapter-Based Framework for Multi-Task Cross-Lingual Transfer (EMNLP, 2020) [paper[180]]
Masking as an Efficient Alternative to Finetuning for Pretrained Language Models (EMNLP, 2020) [paper[181]] [code[182]]
Effcient Continuous Pareto Exploration in Multi-Task Learning (ICML, 2020) [paper[183]] [code[184]]
Which Tasks Should Be Learned Together in Multi-task Learning? (ICML, 2020) [paper[185]] [code[186]]
Learning to Branch for Multi-Task Learning (ICML, 2020) [paper[187]]
Partly Supervised Multitask Learning (ICMLA, 2020) paper[188]
Understanding and Improving Information Transfer in Multi-Task Learning (ICLR, 2020) [paper[189]]
Measuring and Harnessing Transference in Multi-Task Learning (arXiv, 2020) [paper[190]]
Multi-Task Semi-Supervised Adversarial Autoencoding for Speech Emotion Recognition (arXiv, 2020) [paper[191]]
Learning Sparse Sharing Architectures for Multiple Tasks (AAAI, 2020) [paper[192]]
AdapterFusion: Non-Destructive Task Composition for Transfer Learning (arXiv, 2020) [paper[193]]
2019
Adaptive Auxiliary Task Weighting for Reinforcement Learning (NeurIPS, 2019) [paper[194]]
Pareto Multi-Task Learning (NeurIPS, 2019) [paper[195]] [code[196]]
Modular Universal Reparameterization: Deep Multi-task Learning Across Diverse Domains (NeurIPS, 2019) [paper[197]]
Fast and Flexible Multi-Task Classification Using Conditional Neural Adaptive Processes (NeurIPS, 2019) [paper[198]] [code[199]]
[Orthogonal] Regularizing Deep Multi-Task Networks using Orthogonal Gradients (arXiv, 2019) [paper[200]]
Many Task Learning With Task Routing (ICCV, 2019) [paper[201]] [code[202]]
Stochastic Filter Groups for Multi-Task CNNs: Learning Specialist and Generalist Convolution Kernels (ICCV, 2019) [paper[203]]
Deep Elastic Networks with Model Selection for Multi-Task Learning (ICCV, 2019) [paper[204]] [code[205]]
Feature Partitioning for Efficient Multi-Task Architectures (arXiv, 2019) [paper[206]] [code[207]]
Task Selection Policies for Multitask Learning (arXiv, 2019) [paper[208]]
BAM! Born-Again Multi-Task Networks for Natural Language Understanding (ACL, 2019) [paper[209]] [code[210]]
OmniNet: A unified architecture for multi-modal multi-task learning (arXiv, 2019) [paper[211]]
NDDR-CNN: Layerwise Feature Fusing in Multi-Task CNNs by Neural Discriminative Dimensionality Reduction (CVPR, 2019) [paper[212]] [code[213]]
[MTAN + DWA] End-to-End Multi-Task Learning with Attention (CVPR, 2019) [paper[214]] [code[215]]
Attentive Single-Tasking of Multiple Tasks (CVPR, 2019) [paper[216]] [code[217]]
Pattern-Affinitive Propagation Across Depth, Surface Normal and Semantic Segmentation (CVPR, 2019) [paper[218]]
Representation Similarity Analysis for Efficient Task Taxonomy & Transfer Learning (CVPR, 2019) [paper[219]] [code[220]]
[Geometric Loss Strategy (GLS)] MultiNet++: Multi-Stream Feature Aggregation and Geometric Loss Strategy for Multi-Task Learning (CVPR Workshop, 2019) [paper[221]]
Parameter-Efficient Transfer Learning for NLP (ICML, 2019) [paper[222]]
BERT and PALs: Projected Attention Layers for Efficient Adaptation in Multi-Task Learning (ICML, 2019) [paper[223]] [code[224]]
Tasks Without Borders: A New Approach to Online Multi-Task Learning (ICML Workshop, 2019) [paper[225]]
AutoSeM: Automatic Task Selection and Mixing in Multi-Task Learning (NACCL, 2019) [paper[226]] [code[227]]
Multi-Task Deep Reinforcement Learning with PopArt (AAAI, 2019) [paper[228]]
SNR: Sub-Network Routing for Flexible Parameter Sharing in Multi-Task Learning (AAAI, 2019) [paper[229]]
Latent Multi-task Architecture Learning (AAAI, 2019) [paper[230]] [[code](https://github.com/ sebastianruder/sluice-networks "[code")]
Multi-Task Deep Neural Networks for Natural Language Understanding (ACL, 2019) [paper[231]]
2018
Learning to Multitask (NeurIPS, 2018) [paper[232]]
[MGDA] Multi-Task Learning as Multi-Objective Optimization (NeurIPS, 2018) [paper[233]] [code[234]]
Adapting Auxiliary Losses Using Gradient Similarity (arXiv, 2018) [paper[235]] [code[236]]
Piggyback: Adapting a Single Network to Multiple Tasks by Learning to Mask Weights (ECCV, 2018) [paper[237]] [code[238]]
Dynamic Task Prioritization for Multitask Learning (ECCV, 2018) [paper[239]]
A Modulation Module for Multi-task Learning with Applications in Image Retrieval (ECCV, 2018) [paper[240]]
Modeling Task Relationships in Multi-task Learning with Multi-gate Mixture-of-Experts (KDD, 2018) [paper[241]]
Unifying and Merging Well-trained Deep Neural Networks for Inference Stage (IJCAI, 2018) [paper[242]] [code[243]]
Efficient Parametrization of Multi-domain Deep Neural Networks (CVPR, 2018) [paper[244]] [code[245]]
PAD-Net: Multi-tasks Guided Prediction-and-Distillation Network for Simultaneous Depth Estimation and Scene Parsing (CVPR, 2018) [paper[246]]
NestedNet: Learning Nested Sparse Structures in Deep Neural Networks (CVPR, 2018) [paper[247]]
PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruning (CVPR, 2018) [paper[248]] [code[249]]
[Uncertainty] Multi-task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics (CVPR, 2018) [paper[250]]
Deep Asymmetric Multi-task Feature Learning (ICML, 2018) [paper[251]]
[GradNorm] GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networks (ICML, 2018) [paper[252]]
Pseudo-task Augmentation: From Deep Multitask Learning to Intratask Sharing---and Back (ICML, 2018) [paper[253]]
Gradient Adversarial Training of Neural Networks (arXiv, 2018) [paper[254]]
Auxiliary Tasks in Multi-task Learning (arXiv, 2018) [paper[255]]
Routing Networks: Adaptive Selection of Non-linear Functions for Multi-Task Learning (ICLR, 2018) [paper[256]] [code[257]
Beyond Shared Hierarchies: Deep Multitask Learning through Soft Layer Ordering (ICLR, 2018) [paper[258]]
2017
Learning multiple visual domains with residual adapters (NeurIPS, 2017) [paper[259]] [code[260]]
Learning Multiple Tasks with Multilinear Relationship Networks (NeurIPS, 2017) [paper[261]] [code[262]]
Federated Multi-Task Learning (NeurIPS, 2017) [paper[263]] [code[264]]
Multi-task Self-Supervised Visual Learning (ICCV, 2017) [paper[265]]
Adversarial Multi-task Learning for Text Classification (ACL, 2017) [paper[266]]
UberNet: Training a Universal Convolutional Neural Network for Low-, Mid-, and High-Level Vision Using Diverse Datasets and Limited Memory (CVPR, 2017) [paper[267]]
Fully-adaptive Feature Sharing in Multi-Task Networks with Applications in Person Attribute Classification (CVPR, 2017) [paper[268]]
Modular Multitask Reinforcement Learning with Policy Sketches (ICML, 2017) [paper[269]] [code[270]]
SplitNet: Learning to Semantically Split Deep Networks for Parameter Reduction and Model Parallelization (ICML, 2017) [paper[271]] [code[272]]
One Model To Learn Them All (arXiv, 2017) [paper[273]] [code[274]]
[AdaLoss] Learning Anytime Predictions in Neural Networks via Adaptive Loss Balancing (arXiv, 2017) [paper[275]]
Deep Multi-task Representation Learning: A Tensor Factorisation Approach (ICLR, 2017) [paper[276]] [code[277]]
Trace Norm Regularised Deep Multi-Task Learning (ICLR Workshop, 2017) [paper[278]] [code[279]]
When is multitask learning effective? Semantic sequence prediction under varying data conditions (EACL, 2017) [paper[280]] [code[281]]
Identifying beneficial task relations for multi-task learning in deep neural networks (EACL, 2017) [paper[282]]
PathNet: Evolution Channels Gradient Descent in Super Neural Networks (arXiv, 2017) [paper[283]] [code[284]]
Attributes for Improved Attributes: A Multi-Task Network Utilizing Implicit and Explicit Relationships for Facial Attribute Classification (AAAI, 2017) [paper[285]]
2016 and earlier
Learning values across many orders of magnitude (NeurIPS, 2016) [paper[286]]
Integrated Perception with Recurrent Multi-Task Neural Networks (NeurIPS, 2016) [paper[287]]
Unifying Multi-Domain Multi-Task Learning: Tensor and Neural Network Perspectives (arXiv, 2016) [paper[288]]
Progressive Neural Networks (arXiv, 2016) [paper[289]]
Deep multi-task learning with low level tasks supervised at lower layers (ACL, 2016) [paper[290]]
[Cross-Stitch] Cross-Stitch Networks for Multi-task Learning (CVPR,2016) [paper[291]] [code[292]]
Asymmetric Multi-task Learning based on Task Relatedness and Confidence (ICML, 2016) [paper[293]]
MultiNet: Real-time Joint Semantic Reasoning for Autonomous Driving (arXiv, 2016) [paper[294]] [code[295]]
A Unified Perspective on Multi-Domain and Multi-Task Learning (ICLR, 2015) [paper[296]]
Facial Landmark Detection by Deep Multi-task Learning (ECCV, 2014) [paper[297]] [code[298]]
Learning Task Grouping and Overlap in Multi-task Learning (ICML, 2012) [paper[299]]
Learning with Whom to Share in Multi-task Feature Learning (ICML, 2011) [paper[300]]
Semi-Supervised Multi-Task Learning with Task Regularizations (ICDM, 2009) [paper[301]]
Semi-Supervised Multitask Learning (NeurIPS, 2008) [paper[302]]
Multitask Learning (1997) [paper[303]]
Awesome Multi-domain Multi-task Learning[304]
Workshops
Workshop on Multi-Task Learning in Computer Vision (DeepMTL) at ICCV 2021[305]
Adaptive and Multitask Learning: Algorithms & Systems Workshop (AMTL) at ICML 2019[306]
Workshop on Multi-Task and Lifelong Reinforcement Learning at ICML 2015[307]
Transfer and Multi-Task Learning: Trends and New Perspectives at NeurIPS 2015[308]
Second Workshop on Transfer and Multi-task Learning at NeurIPS 2014[309]
New Directions in Transfer and Multi-Task: Learning Across Domains and Tasks Workshop at NeurIPS 2013[310]
Online Courses
CS 330: Deep Multi-Task and Meta Learning[311]
Related awesome list
https://github.com/SimonVandenhende/Awesome-Multi-Task-Learning
https://github.com/Manchery/awesome-multi-task-learning
想和你一起学习进步！『NewBeeNLP』目前已经建立了多个不同方向交流群（机器学习 / 深度学习 / 自然语言处理 / 搜索推荐 / 图网络 / 面试交流 / 等），名额有限，赶紧添加下方微信加入一起讨论交流吧！（注意一定o要备注信息才能通过）
参考资料
[paper: https://arxiv.org/pdf/2103.13318.pdf
[2][paper: https://arxiv.org/abs/2004.13379
[3][code: https://github.com/SimonVandenhende/Multi-Task-Learning-PyTorch
[4][paper: https://ieeexplore.ieee.org/abstract/document/9392366
[5][paper: http://arxiv.org/abs/2009.09796
[6]best paper]) [[paper: https://openaccess.thecvf.com/content_cvpr_2018/papers/Zamir_Taskonomy_Disentangling_Task_CVPR_2018_paper.pdf
[7][dataset: http://taskonomy.stanford.edu/
[8][paper: https://ieeexplore.ieee.org/document/8848395
[9][paper: http://arxiv.org/abs/1706.05098
[10]NYUv2]** Indoor Segmentation and Support Inference from RGBD Images (ECCV, 2012) [[paper: https://cs.nyu.edu/~silberman/papers/indoor_seg_support.pdf
[11][dataset: https://cs.nyu.edu/~silberman/datasets/nyu_depth_v2.html
[12]Cityscapes]** The Cityscapes Dataset for Semantic Urban Scene Understanding (CVPR, 2016) [[paper: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7780719
[13][dataset: https://www.cityscapes-dataset.com/
[14]PASCAL-Context]** The Role of Context for Object Detection and Semantic Segmentation in the Wild (CVPR, 2014) [[paper: https://cs.stanford.edu/~roozbeh/pascal-context/mottaghi_et_al_cvpr14.pdf
[15][dataset: https://cs.stanford.edu/~roozbeh/pascal-context/
[16]Taskonomy]** Taskonomy: Disentangling Task Transfer Learning (CVPR, 2018 [best paper]) [[paper: https://openaccess.thecvf.com/content_cvpr_2018/papers/Zamir_Taskonomy_Disentangling_Task_CVPR_2018_paper.pdf
[17][dataset: http://taskonomy.stanford.edu/
[18]KITTI]** Vision meets robotics: The KITTI dataset (IJRR, 2013) [[paper: http://www.cvlibs.net/publications/Geiger2013IJRR.pdf
[19]dataset: http://www.cvlibs.net/datasets/kitti/
[20]SUN RGB-D]** SUN RGB-D: A RGB-D Scene Understanding Benchmark Suite (CVPR 2015) [[paper: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7298655
[21][dataset: https://rgbd.cs.princeton.edu
[22]BDD100K]** BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning (CVPR, 2020) [[paper: https://openaccess.thecvf.com/content_CVPR_2020/papers/Yu_BDD100K_A_Diverse_Driving_Dataset_for_Heterogeneous_Multitask_Learning_CVPR_2020_paper.pdf
[23][dataset: https://bdd-data.berkeley.edu/
[24]Omnidata]** Omnidata: A Scalable Pipeline for Making Multi-Task Mid-Level Vision Datasets from 3D Scans (ICCV, 2021) [[paper: https://arxiv.org/pdf/2110.04994.pdf
[25][project: https://omnidata.vision
[26]Meta-dataset]** Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples (ICLR, 2020) [[paper: https://openreview.net/pdf?id=rkgAGAVKPr
[27][dataset: https://github.com/google-research/meta-dataset
[28]Visual Domain Decathlon]** Learning multiple visual domains with residual adapters (NeurIPS, 2017) [[paper: https://arxiv.org/abs/1705.08045
[29][dataset: https://www.robots.ox.ac.uk/~vgg/decathlon/
[30]CelebA]** Deep Learning Face Attributes in the Wild (ICCV, 2015) [[paper: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7410782
[31][dataset: http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html
[32][Multi-Task-Learning-PyTorch: https://github.com/SimonVandenhende/Multi-Task-Learning-PyTorch
[33][Auto-λ: https://github.com/lorenmt/auto-lambda
[34][UniversalRepresentations: https://github.com/VICO-UoE/UniversalRepresentations
[35]Multi-task Dense Prediction: https://github.com/VICO-UoE/UniversalRepresentations/tree/main/DensePred
[36]Multi-domain Classification: https://github.com/VICO-UoE/UniversalRepresentations/tree/main/VisualDecathlon
[37]Cross-domain Few-shot Learning: https://github.com/VICO-UoE/URL
[38][MTAN: https://github.com/lorenmt/mtan
[39][ASTMT: https://github.com/facebookresearch/astmt
[40][LibMTL: https://github.com/median-research-group/libmtl
[41][MTPSL: https://github.com/VICO-UoE/MTPSL
[42][Resisual Adapater: https://github.com/srebuffi/residual_adapters
[43]Auto-λ]** Auto-λ: Disentangling Dynamic Task Relationships (TMLR, 2022) [[paper: https://arxiv.org/pdf/2202.03091.pdf
[44][code: https://github.com/lorenmt/auto-lambda
[45]Universal Representations]** Universal Representations: A Unified Look at Multiple Task and Domain Learning (arXiv, 2022) [[paper: https://arxiv.org/pdf/2204.02744.pdf
[46][code: https://github.com/VICO-UoE/UniversalRepresentations
[47][paper: https://arxiv.org/abs/2207.03036
[48][code: https://github.com/TencentARC/SFDA
[49][paper: https://arxiv.org/abs/2207.03337
[50][code: https://github.com/Adamdad/KnowledgeFactor
[51]InvPT]** Inverted Pyramid Multi-task Transformer for Dense Scene Understanding (ECCV, 2022) [[paper: https://arxiv.org/pdf/2203.07997.pdf
[52][code: https://github.com/prismformore/InvPT
[53]MultiMAE]** MultiMAE: Multi-modal Multi-task Masked Autoencoders (ECCV, 2022) [[paper: https://arxiv.org/pdf/2204.01678.pdf
[54][code: https://multimae.epfl.ch
[55][paper: https://proceedings.mlr.press/v162/momma22a.html
[56][paper: https://proceedings.mlr.press/v162/javaloy22a.html
[57][paper: https://proceedings.mlr.press/v162/chen22j.html
[58][paper: https://proceedings.mlr.press/v162/bao22c.html
[59][code: https://github.com/zpbao/multi-task-oriented_generative_modeling
[60][paper: https://proceedings.mlr.press/v162/navon22a.html
[61][code: https://github.com/AvivNavon/nash-mtl
[62][paper: https://arxiv.org/pdf/2205.14354.pdf
[63]Gato]** A Generalist Agent (arXiv, 2022) [[paper: https://arxiv.org/pdf/2205.06175.pdf
[64]MTPSL]** Learning Multiple Dense Prediction Tasks from Partially Annotated Data (CVPR, 2022) [[paper: https://arxiv.org/pdf/2111.14893.pdf
[65][code: https://github.com/VICO-UoE/MTPSL
[66]TSA]** Cross-domain Few-shot Learning with Task-specific Adapters (CVPR, 2022) [[paper: https://arxiv.org/pdf/2107.00358.pdf
[67][code: https://github.com/VICO-UoE/URL
[68]OMNIVORE]** OMNIVORE: A Single Model for Many Visual Modalities (CVPR, 2022) [[paper: https://arxiv.org/pdf/2201.08377.pdf
[69][code: https://github.com/facebookresearch/omnivore
[70][paper: https://arxiv.org/pdf/2203.16708.pdf
[71][paper: https://arxiv.org/pdf/2203.14949.pdf
[72][code: https://www.nec-labs.com/~mas/DYMU/
[73]SHIFT]** SHIFT: A Synthetic Driving Dataset for Continuous Multi-Task Domain Adaptation (CVPR, 2022) [[paper: https://openaccess.thecvf.com/content/CVPR2022/papers/Sun_SHIFT_A_Synthetic_Driving_Dataset_for_Continuous_Multi-Task_Domain_Adaptation_CVPR_2022_paper.pdf
[74][code: https://www.vis.xyz/shift/
[75][paper: https://openaccess.thecvf.com/content/CVPR2022/papers/Sun_DiSparse_Disentangled_Sparsification_for_Multitask_Model_Compression_CVPR_2022_paper.pdf
[76][code: https://github.com/SHI-Labs/DiSparse-Multitask-Model-Compression
[77]MulT]** MulT: An End-to-End Multitask Learning Transformer (CVPR, 2022) [[paper: https://openaccess.thecvf.com/content/CVPR2022/papers/Bhattacharjee_MulT_An_End-to-End_Multitask_Learning_Transformer_CVPR_2022_paper.pdf
[78][code: https://github.com/IVRL/MulT
[79][paper: https://openaccess.thecvf.com/content/CVPR2022/papers/Vasudevan_Sound_and_Visual_Representation_Learning_With_Multiple_Pretraining_Tasks_CVPR_2022_paper.pdf
[80][paper: https://arxiv.org/abs/2204.05698
[81][paper: https://arxiv.org/pdf/2205.12755.pdf
[82][code: https://github.com/google-research/google-research/tree/master/muNet
[83][paper: https://arxiv.org/pdf/2202.13914.pdf
[84][paper: https://openreview.net/pdf?id=kG0AtPi6JI1
[85][paper: https://openreview.net/pdf?id=8H5bpVwvt5
[86][code: https://github.com/Adaptive-RL/AdaRL-code
[87][paper: https://openreview.net/pdf?id=0RDcd5Axok
[88][code: https://github.com/jxhe/unify-parameter-efficient-tuning
[89]Rotograd]** Rotograd: Dynamic Gradient Homogenization for Multi-Task Learning (ICLR, 2022) [[paper: https://openreview.net/pdf?id=T8wHz4rnuGL
[90][code: https://github.com/adrianjav/rotograd
[91][paper: https://openreview.net/pdf?id=8Py-W8lSUgy
[92][paper: https://openreview.net/pdf?id=ltM1RMZntpu
[93][code: https://github.com/CogComp/TAWT
[94][paper: https://openaccess.thecvf.com/content/WACV2022/papers/Wang_Semi-Supervised_Multi-Task_Learning_for_Semantics_and_Depth_WACV_2022_paper.pdf
[95][paper: https://arxiv.org/pdf/2201.04122.pdf
[96][paper: http://arxiv.org/abs/2109.04617
[97]CAGrad]** Conflict-Averse Gradient Descent for Multi-task Learning (NeurIPS, 2021) [[paper: https://openreview.net/pdf?id=61Qh8tULj
[98][code: https://github.com/Cranial-XIX/CAGrad
[99][paper: https://arxiv.org/pdf/2111.10603.pdf
[100][paper: http://arxiv.org/abs/2104.13874
[101][code: https://github.com/brdav/atrc
[102][paper: https://openaccess.thecvf.com/content/ICCV2021/papers/Ghiasi_Multi-Task_Self-Training_for_Learning_General_Representations_ICCV_2021_paper.pdf
[103][paper: https://openaccess.thecvf.com/content/ICCV2021/html/Sun_Task_Switching_Network_for_Multi-Task_Learning_ICCV_2021_paper.html
[104][code: https://github.com/GuoleiSun/TSNs
[105][paper: https://arxiv.org/pdf/2110.04994.pdf
[106][project: https://omnidata.vision
[107][paper: https://arxiv.org/abs/2103.10919
[108][code: https://github.com/EPFL-VILAB/XDEnsembles
[109][paper: https://openaccess.thecvf.com/content/ICCV2021/papers/Wang_Domain_Adaptive_Semantic_Segmentation_With_Self-Supervised_Depth_Estimation_ICCV_2021_paper.pdf
[110][code: https://qin.ee/corda
[111]URL]** Universal Representation Learning from Multiple Domains for Few-shot Classification (ICCV, 2021) [[paper: https://openaccess.thecvf.com/content/ICCV2021/papers/Li_Universal_Representation_Learning_From_Multiple_Domains_for_Few-Shot_Classification_ICCV_2021_paper.pdf
[112][code: https://github.com/VICO-UoE/URL
[113]tri-M]** A Multi-Mode Modulator for Multi-Domain Few-Shot Classification (ICCV, 2021) [[paper: https://openaccess.thecvf.com/content/ICCV2021/papers/Liu_A_Multi-Mode_Modulator_for_Multi-Domain_Few-Shot_Classification_ICCV_2021_paper.pdf
[114][code: https://github.com/csyanbin/tri-M-ICCV
[115][paper: https://openaccess.thecvf.com/content/ICCV2021W/ERCVAD/papers/Heuer_MultiTask-CenterNet_MCN_Efficient_and_Diverse_Multitask_Learning_Using_an_Anchor_ICCVW_2021_paper.pdf
[116][paper: https://arxiv.org/pdf/2110.02549.pdf
[117][paper: https://www.cinc.org/2021/Program/accepted/115_Preprint.pdf
[118][code: https://github.com/SMKamrulHasan/MTCTL
[119][paper: http://arxiv.org/abs/2102.06177
[120]FLUTE]** Learning a Universal Template for Few-shot Dataset Generalization (ICML, 2021) [[paper: https://arxiv.org/pdf/2105.07029.pdf
[121][code: https://github.com/google-research/meta-dataset
[122][paper: http://arxiv.org/abs/2110.04366
[123][paper: http://arxiv.org/abs/2102.10772
[124][paper: https://openaccess.thecvf.com/content/CVPR2021/papers/Saha_Learning_To_Relate_Depth_and_Semantics_for_Unsupervised_Domain_Adaptation_CVPR_2021_paper.pdf
[125][code: https://github.com/susaha/ctrl-uda
[126][paper: https://openaccess.thecvf.com/content/CVPR2021/html/Popovic_CompositeTasking_Understanding_Images_by_Spatial_Composition_of_Tasks_CVPR_2021_paper.html
[127][code: https://github.com/nikola3794/composite-tasking
[128][paper: https://openaccess.thecvf.com/content/CVPR2021/papers/Georgescu_Anomaly_Detection_in_Video_via_Self-Supervised_and_Multi-Task_Learning_CVPR_2021_paper.pdf
[129][paper: https://openaccess.thecvf.com/content/CVPR2021/papers/Lu_Taskology_Utilizing_Task_Relations_at_Scale_CVPR_2021_paper.pdf
[130][paper: https://arxiv.org/pdf/2012.10782.pdf
[131][code: https://github.com/lhoyer/improving_segmentation_with_selfsupervised_depth
[132][paper: https://arxiv.org/pdf/2108.12545.pdf
[133][code: https://github.com/lhoyer/improving_segmentation_with_selfsupervised_depth
[134][paper: https://aclanthology.org/2021.findings-emnlp.240
[135][paper: https://openreview.net/forum?id=de11dbHzAMF
[136][code: https://github.com/CAMTL/CA-MTL
[137]Gradient Vaccine]** Gradient Vaccine: Investigating and Improving Multi-task Optimization in Massively Multilingual Models (ICLR, 2021) [[paper: https://openreview.net/forum?id=F1vEjWK-lH_
[138]IMTL]** Towards Impartial Multi-task Learning (ICLR, 2021) [[paper: https://openreview.net/forum?id=IMPnRXEWpvr
[139][paper: https://openreview.net/forum?id=Cri3xz59ga
[140]URT]** A Universal Representation Transformer Layer for Few-Shot Image Classification (ICLR, 2021) [[paper: https://arxiv.org/pdf/2006.11702.pdf
[141][code: https://github.com/liulu112601/URT
[142][paper: http://arxiv.org/abs/1910.04915
[143][paper: https://openaccess.thecvf.com/content/WACV2021/papers/Groenendijk_Multi-Loss_Weighting_With_Coefficient_of_Variations_WACV_2021_paper.pdf
[144][code: https://github.com/rickgroen/cov-weighting
[145][paper: http://arxiv.org/abs/2003.13661
[146][code: https://github.com/RchalYang/Soft-Module
[147][paper: http://arxiv.org/abs/1911.12423
[148][code: https://github.com/sunxm2357/AdaShare
[149]GradDrop]** Just Pick a Sign: Optimizing Deep Multitask Models with Gradient Sign Dropout (NeurIPS, 2020) [[paper: https://proceedings.NeurIPS.cc//paper/2020/file/16002f7a455a94aa4e91cc34ebdb9f2d-Paper.pdf
[150][code: https://github.com/tensorflow/lingvo/blob/master/lingvo/core/graddrop.py
[151]PCGrad]** Gradient Surgery for Multi-Task Learning (NeurIPS, 2020) [[paper: http://arxiv.org/abs/2001.06782
[152][tensorflow: https://github.com/tianheyu927/PCGrad
[153][pytorch: https://github.com/WeiChengTseng/Pytorch-PCGrad
[154][paper: https://proceedings.NeurIPS.cc//paper/2020/file/59587bffec1c7846f3e34230141556ae-Paper.pdf
[155][paper: https://www.aclweb.org/anthology/2020.wmt-1.72/
[156][paper: http://arxiv.org/abs/2011.09824
[157][paper: http://arxiv.org/abs/2008.10292
[158][code: https://github.com/brdav/bmtas
[159][paper: http://arxiv.org/abs/1904.02920
[160][paper: http://arxiv.org/abs/2001.06902
[161][code: https://github.com/SimonVandenhende/Multi-Task-Learning-PyTorch
[162][paper: http://arxiv.org/abs/2007.12540
[163][code: https://github.com/menelaoskanakis/RCM
[164][paper: https://arxiv.org/pdf/2003.09338.pdf
[165][code: https://github.com/dvornikita/SUR
[166][paper: https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123470154.pdf
[167][code: https://github.com/columbia/MTRobust
[168][paper: https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123710494.pdf
[169][code: https://github.com/cvai-repo/duality-diagram-similarity
[170]KD4MTL]** Knowledge Distillation for Multi-task Learning (ECCV Workshop) [[paper: https://arxiv.org/pdf/2007.06889.pdf
[171][code: https://github.com/VICO-UoE/KD4MTL
[172][paper: https://arxiv.org/abs/2003.14058
[173][code: https://github.com/bhpfelix/MTLNAS
[174][paper: https://consistency.epfl.ch/Cross_Task_Consistency_CVPR2020.pdf
[175][code: https://github.com/EPFL-VILAB/XTConsistency
[176]paper: https://openaccess.thecvf.com/content_CVPR_2020/papers/Lu_12-in-1_Multi-Task_Vision_and_Language_Representation_Learning_CVPR_2020_paper.pdf
[177][code: https://github.com/facebookresearch/vilbert-multi-task
[178][paper: https://openaccess.thecvf.com/content_CVPR_2020/papers/Chen_A_Multi-Task_Mean_Teacher_for_Semi-Supervised_Shadow_Detection_CVPR_2020_paper.pdf
[179][code: https://github.com/eraserNut/MTMT
[180][paper: https://doi.org/10.18653/v1/2020.emnlp-main.617
[181][paper: http://arxiv.org/abs/2004.12406
[182][code: https://github.com/ptlmasking/maskbert
[183][paper: http://proceedings.mlr.press/v119/ma20a/ma20a.pdf
[184][code: https://github.com/mit-gfx/ContinuousParetoMTL
[185][paper: http://arxiv.org/abs/1905.07553
[186][code: https://github.com/tstandley/taskgrouping
[187][paper: https://arxiv.org/abs/2006.01895
[188]paper: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9356271
[189][paper: https://arxiv.org/abs/2005.00944
[190][paper: https://arxiv.org/abs/2010.15413
[191][paper: https://arxiv.org/pdf/1907.06078.pdf
[192][paper: http://arxiv.org/abs/1911.05034
[193][paper: http://arxiv.org/abs/2005.00247
[194][paper: https://papers.nips.cc/paper/2019/hash/0e900ad84f63618452210ab8baae0218-Abstract.html
[195][paper: http://papers.nips.cc/paper/9374-pareto-multi-task-learning.pdf
[196][code: https://github.com/Xi-L/ParetoMTL
[197][paper: http://arxiv.org/abs/1906.00097
[198][paper: https://github.com/cambridge-mlg/cnaps
[199][code: https://proceedings.neurips.cc/paper/2019/file/1138d90ef0a0848a542e57d1595f58ea-Paper.pdf
[200]Orthogonal]** Regularizing Deep Multi-Task Networks using Orthogonal Gradients (arXiv, 2019) [[paper: http://arxiv.org/abs/1912.06844
[201][paper: https://openaccess.thecvf.com/content_ICCV_2019/papers/Strezoski_Many_Task_Learning_With_Task_Routing_ICCV_2019_paper.pdf
[202][code: https://github.com/gstrezoski/TaskRouting
[203][paper: https://arxiv.org/abs/1908.09597
[204][paper: http://arxiv.org/abs/1909.04860
[205][code: https://github.com/rllab-snu/Deep-Elastic-Network
[206][paper: https://arxiv.org/abs/1908.04339
[207][code: https://github.com/google/multi-task-architecture-search
[208][paper: http://arxiv.org/abs/1907.06214
[209][paper: https://www.aclweb.org/anthology/P19-1595/
[210][code: https://github.com/google-research/google-research/tree/master/bam
[211][paper: http://arxiv.org/abs/1907.07804
[212][paper: https://arxiv.org/abs/1801.08297
[213][code: https://github.com/ethanygao/NDDR-CNN
[214]MTAN + DWA]** End-to-End Multi-Task Learning with Attention (CVPR, 2019) [[paper: http://arxiv.org/abs/1803.10704
[215][code: https://github.com/lorenmt/mtan
[216][paper: http://arxiv.org/abs/1904.08918
[217][code: https://github.com/facebookresearch/astmt
[218][paper: https://openaccess.thecvf.com/content_CVPR_2019/papers/Zhang_Pattern-Affinitive_Propagation_Across_Depth_Surface_Normal_and_Semantic_Segmentation_CVPR_2019_paper.pdf
[219][paper: https://arxiv.org/abs/1904.11740
[220][code: https://github.com/kshitijd20/RSA-CVPR19-release
[221]Geometric Loss Strategy (GLS)]** MultiNet++: Multi-Stream Feature Aggregation and Geometric Loss Strategy for Multi-Task Learning (CVPR Workshop, 2019) [[paper: http://arxiv.org/abs/1904.08492
[222][paper: http://arxiv.org/abs/1902.00751
[223][paper: http://arxiv.org/abs/1902.02671
[224][code: https://github.com/AsaCooperStickland/Bert-n-Pals
[225][paper: https://openreview.net/pdf?id=HkllV5Bs24
[226][paper: https://arxiv.org/abs/1904.04153
[227][code: https://github.com/HanGuo97/AutoSeM
[228][paper: https://doi.org/10.1609/aaai.v33i01.33013796
[229][paper: https://ojs.aaai.org/index.php/AAAI/article/view/3788/3666
[230][paper: https://arxiv.org/abs/1705.08142
[231][paper: https://arxiv.org/pdf/1901.11504.pdf
[232][paper: https://papers.nips.cc/paper/2018/file/aeefb050911334869a7a5d9e4d0e1689-Paper.pdf
[233]MGDA]** Multi-Task Learning as Multi-Objective Optimization (NeurIPS, 2018) [[paper: http://arxiv.org/abs/1810.04650
[234][code: https://github.com/isl-org/MultiObjectiveOptimization
[235][paper: http://arxiv.org/abs/1812.02224
[236][code: https://github.com/szkocot/Adapting-Auxiliary-Losses-Using-Gradient-Similarity
[237][paper: https://openaccess.thecvf.com/content_ECCV_2018/papers/Arun_Mallya_Piggyback_Adapting_a_ECCV_2018_paper.pdf
[238][code: https://github.com/arunmallya/piggyback
[239][paper: https://openaccess.thecvf.com/content_ECCV_2018/papers/Michelle_Guo_Focus_on_the_ECCV_2018_paper.pdf
[240][paper: https://arxiv.org/abs/1807.06708
[241][paper: https://dl.acm.org/doi/pdf/10.1145/3219819.3220007
[242][paper: http://arxiv.org/abs/1805.04980
[243][code: https://github.com/ivclab/NeuralMerger
[244][paper: https://openaccess.thecvf.com/content_cvpr_2018/papers/Rebuffi_Efficient_Parametrization_of_CVPR_2018_paper.pdf
[245][code: https://github.com/srebuffi/residual_adapters
[246][paper: https://openaccess.thecvf.com/content_cvpr_2018/papers/Xu_PAD-Net_Multi-Tasks_Guided_CVPR_2018_paper.pdf
[247][paper: https://openaccess.thecvf.com/content_cvpr_2018/papers/Kim_NestedNet_Learning_Nested_CVPR_2018_paper.pdf
[248][paper: https://openaccess.thecvf.com/content_cvpr_2018/papers/Mallya_PackNet_Adding_Multiple_CVPR_2018_paper.pdf
[249][code: https://github.com/arunmallya/packnet
[250]Uncertainty]** Multi-task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics (CVPR, 2018) [[paper: https://openaccess.thecvf.com/content_cvpr_2018/papers/Kendall_Multi-Task_Learning_Using_CVPR_2018_paper.pdf
[251][paper: http://proceedings.mlr.press/v80/lee18d/lee18d.pdf
[252]GradNorm]** GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networks (ICML, 2018) [[paper: http://arxiv.org/abs/1711.02257
[253][paper: http://arxiv.org/abs/1803.04062
[254][paper: http://arxiv.org/abs/1806.08028
[255][paper: http://arxiv.org/abs/1805.06334
[256][paper: http://arxiv.org/abs/1711.01239
[257][code: https://github.com/cle-ros/RoutingNetworks
[258][paper: http://arxiv.org/abs/1711.00108
[259][paper: https://papers.nips.cc/paper/2017/file/e7b24b112a44fdd9ee93bdf998c6ca0e-Paper.pdf
[260][code: https://github.com/srebuffi/residual_adapters
[261][paper: https://proceedings.NeurIPS.cc/paper/2017/file/03e0704b5690a2dee1861dc3ad3316c9-Paper.pdf
[262][code: https://github.com/thuml/MTlearn
[263][paper: https://proceedings.NeurIPS.cc/paper/2017/file/6211080fa89981f66b1a0c9d55c61d0f-Paper.pdf
[264][code: https://github.com/gingsmith/fmtl
[265][paper: http://arxiv.org/abs/1708.07860
[266][paper: http://arxiv.org/abs/1704.05742
[267][paper: https://arxiv.org/abs/1609.02132
[268][paper: https://openaccess.thecvf.com/content_cvpr_2017/papers/Lu_Fully-Adaptive_Feature_Sharing_CVPR_2017_paper.pdf
[269][paper: http://arxiv.org/abs/1611.01796
[270][code: https://github.com/jacobandreas/psketch
[271][paper: http://proceedings.mlr.press/v70/kim17b.html
[272][code: https://github.com/dalgu90/splitnet-wrn
[273][paper: http://arxiv.org/abs/1706.05137
[274][code: https://github.com/tensorflow/tensor2tensor
[275]AdaLoss]** Learning Anytime Predictions in Neural Networks via Adaptive Loss Balancing (arXiv, 2017) [[paper: http://arxiv.org/abs/1708.06832
[276][paper: https://arxiv.org/abs/1605.06391
[277][code: https://github.com/wOOL/DMTRL
[278][paper: http://arxiv.org/abs/1606.04038
[279][code: https://github.com/wOOL/TNRDMTL
[280][paper: http://arxiv.org/abs/1612.02251
[281][code: https://github.com/bplank/multitasksemantics
[282][paper: http://arxiv.org/abs/1702.08303
[283][paper: http://arxiv.org/abs/1701.08734
[284][code: https://github.com/jsikyoon/pathnet
[285][paper: https://www.aaai.org/ocs/index.php/AAAI/AAAI17/paper/viewFile/14749/14282
[286][paper: https://arxiv.org/abs/1602.07714
[287][paper: https://proceedings.neurips.cc/paper/2016/file/06409663226af2f3114485aa4e0a23b4-Paper.pdf
[288][paper: http://arxiv.org/abs/1611.09345
[289][paper: https://arxiv.org/abs/1606.04671
[290][paper: https://www.aclweb.org/anthology/P16-2038.pdf
[291]Cross-Stitch]** Cross-Stitch Networks for Multi-task Learning (CVPR,2016) [[paper: https://arxiv.org/abs/1604.03539
[292][code: https://github.com/helloyide/Cross-stitch-Networks-for-Multi-task-Learning
[293][paper: http://proceedings.mlr.press/v48/leeb16.pdf
[294][paper: http://arxiv.org/abs/1612.07695
[295][code: https://github.com/MarvinTeichmann/MultiNet
[296][paper: http://arxiv.org/abs/1412.7489
[297][paper: https://personal.ie.cuhk.edu.hk/~ccloy/files/eccv_2014_deepfacealign.pdf
[298][code: http://mmlab.ie.cuhk.edu.hk/projects/TCDCN.html
[299][paper: http://arxiv.org/abs/1206.6417
[300][paper: http://www.cs.utexas.edu/~grauman/papers/icml2011.pdf
[301][paper: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5360282
[302][paper: https://proceedings.neurips.cc/paper/2007/file/a34bacf839b923770b2c360eefa26748-Paper.pdf
[303][paper: https://link.springer.com/content/pdf/10.1023/A:1007379606734.pdf
[304]Awesome Multi-domain Multi-task Learning: https://github.com/WeiHongLee/Awesome-Multi-Domain-Multi-Task-Learning
[305]Workshop on Multi-Task Learning in Computer Vision (DeepMTL) at ICCV 2021: https://sites.google.com/view/deepmtlworkshop/home
[306]Adaptive and Multitask Learning: Algorithms & Systems Workshop (AMTL) at ICML 2019: https://www.amtl-workshop.org
[307]Workshop on Multi-Task and Lifelong Reinforcement Learning at ICML 2015: https://sites.google.com/view/mtlrl
[308]Transfer and Multi-Task Learning: Trends and New Perspectives at NeurIPS 2015: https://nips.cc/Conferences/2015/Schedule?showEvent=4939
[309]Second Workshop on Transfer and Multi-task Learning at NeurIPS 2014: https://sites.google.com/site/multitaskwsnips2014/
[310]New Directions in Transfer and Multi-Task: Learning Across Domains and Tasks Workshop at NeurIPS 2013: https://sites.google.com/site/learningacross/home
[311]CS 330: Deep Multi-Task and Meta Learning: https://cs330.stanford.edu
COLING'22 | SelfMix：针对带噪数据集的半监督学习方法
ACMMM 2022 | 首个针对跨语言跨模态检索的噪声鲁棒研究工作
微信扫码关注该文公众号作者