Multi-task trade-offs in machine learning can be addressed via Pareto Front Learning (PFL) methods that parameterize the Pareto Front (PF) with a single model. PFL permits to select the desired operational point during inference, contrary to traditional Multi-Task Learning (MTL) that optimizes for a single trade-off decided prior to training. However, recent PFL methodologies suffer from limited scalability, slow convergence, and excessive memory requirements, while exhibiting inconsistent mappings from preference to objective space. We introduce PaLoRA, a novel parameter-efficient method that addresses these limitations in two ways. First, we augment any neural network architecture with task-specific low-rank adapters and continuously parameterize the PF in their convex hull. Our approach steers the original model and the adapters towards learning general and task-specific features, respectively. Second, we propose a deterministic sampling schedule of preference vectors that reinforces this division of labor, enabling faster convergence and strengthening the validity of the mapping from preference to objective space throughout training. Our experiments show that PaLoRA outperforms state-of-the-art MTL and PFL baselines across various datasets, scales to large networks, reducing the memory overhead $23.8-31.7$ times compared with competing PFL baselines in scene understanding benchmarks.
A Single Deep Preference-Conditioned Policy for Learning Pareto Coverage Sets
International Research Center for Neurointelligence, The University of Tokyo（国际神经智能研究中心，东京大学）；ATR Neural Information Analysis Laboratories（ATR神经信息分析实验室）；Department of Information Science, Kyoto University（京都大学信息科学系）
Learning Heterogeneous Performance-Fairness Trade-offs in Federated Learning
Department of Computer Science and Engineering and the Research Institute of Trustworthy Autonomous Systems at Southern University of Science and Technology（计算机科学与工程系和可信自主系统研究 institute at 南方科技大学）
Pareto Continual Learning: Preference-Conditioned Learning and Adaption for Dynamic Stability-Plasticity Trade-off
City University of Hong Kong（香港城市大学）；HK Institute of Science & Innovation, Chinese Academy of Sciences（中国科学院香港创新研究院）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；University of Chinese Academy of Sciences（中国科学院大学）；City University of Hong Kong Shenzhen Research Institute（香港城市大学深圳研究院）；University of Science and Technology of China（中国科学技术大学）
Improving Pareto Front Learning via Multi-Sample Hypernetworks
College of Engineering and Computer Science, VinUniversity（VinUniversity工程与计算机科学学院）；School of Applied Mathematics and Informatics, Hanoi University of Science and Technology（河内科技大学应用数学与信息学院）
Interactive Pareto navigation for deep multi-task learning
Department of Computer Science, TU Dortmund, Dortmund, Germany（多特蒙德工业大学计算机科学系，德国多特蒙德）；Lamarr Institute for Machine Learning and Artificial Intelligence（拉马尔机器学习和人工智能研究所）
多目标优化（MOO）问题是机器学习中普遍存在的问题。这些问题有一组最优解，称为帕累托前沿，其中前沿上的每个点代表可能冲突的目标之间的不同权衡。最近的MOO方法可以在损失空间中瞄准特定的期望射线，然而，大多数方法仍然面临两个严重的限制：（i）必须为前面的每个点训练单独的模型;以及（ii）在优化过程之前必须知道确切的权衡。在这里，我们解决了学习整个帕累托前沿的问题，并能够在训练后在前沿上选择所需的操作点。我们称之为帕累托前沿学习（Pareto Front Learning，PFL）。我们描述了一种使用超网络实现的PFL方法，我们将其称为帕累托超网络（PHN）。PHN使用单个超网络同时学习整个帕累托前沿，该超网络接收所需的偏好向量作为输入，并返回帕累托最优模型，其损失向量在所需的射线中。与训练多个模型相比，统一模型具有运行时效率，并且可以推广到训练期间未使用的新操作点。我们评估我们的方法在广泛的问题，从多任务回归和分类公平。PHN学习整个帕累托前沿的时间与学习前沿上的单个点的时间大致相同，并且同时达到更好的解集。此外，我们还证明了PHN可以扩展以生成像ResNet 18这样的大型模型。PFL为新的应用程序打开了大门，在新的应用程序中，模型是根据仅在运行时间可用的首选项来选择的。
C-MORL: Multi-Objective Reinforcement Learning through Efficient Discovery of Pareto Front
University of Oxford（牛津大学）；The Hong Kong University of Science and Technology（香港科学与技术大学）；Queen Mary University of London（伦敦大学玛丽女王学院）；Microsoft Research Asia（微软亚洲研究院）；Peking University（北京大学）；University of Alberta（阿尔伯塔大学）
Selective Ensemble Based on Preference-Directed Multi-Objective Bandits
Department of Computer Science, Zhejiang Gongshang University（浙江工商大学计算机科学与技术学院）；Center for Advanced Intelligence Project, RIKEN（日本理化学研究所先进智能研究中心）；Graduate School of Frontier Sciences, The University of Tokyo（东京大学新领域创成科学研究科）；National Key Laboratory for Novel Software Technology, Nanjing University（南京大学计算机软件新技术国家重点实验室）；School of Artificial Intelligence, Nanjing University（南京大学人工智能学院）
Divide and Conquer: Provably Unveiling the Pareto Front with Multi-Objective Reinforcement Learning
Vrije Universiteit Brussel（布鲁塞尔自由大学）；Université de Montréal（蒙特利尔大学）；Mila - Quebec AI Institute（米拉-魁北克人工智能研究所）；University of Galway（戈尔韦大学）；City of Amsterdam（阿姆斯特丹市）；Utrecht University（乌得勒支大学）
GraphAllocBench: A Flexible Benchmark for Preference-Conditioned Multi-Objective Policy Learning
University of California, Los Angeles（加州大学洛杉矶分校）；USC Institute for Creative Technologies（美国南加州大学创意技术研究所）；U.S. Army DEVCOM Army Research Laboratory（美国陆军 DEVCOM 军事研究实验室）
Center for Intelligent and Networked System (CFINS)（智能与网络化系统中心）；Department of Automation（自动化系）；Beijing National Research Center for Information Science and Technology（北京信息科学与技术国家研究中心）；Beijing Key Laboratory of Embodied Intelligence Systems（北京具身智能系统重点实验室）；Tsinghua University（清华大学）
Bi-level Multi-objective Evolutionary Learning: A Case Study on Multi-task Graph Neural Topology Search
Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education（教育部智能感知与图像理解重点实验室）；International Research Center for Intelligent Perception（智能感知国际研究中心）
