Auto-Lambda: Disentangling Dynamic Task Relationships
Understanding the structure of multiple related tasks allows for multi-task learning to improve the generalisation ability of one or all of them. However, it usually requires training each pairwise combination of tasks together in order to capture task relationships, at an extremely high computational cost. In this work, we learn task relationships via an automated weighting framework, named Auto-Lambda. Unlike previous methods where task relationships are assumed to be fixed, Auto-Lambda is a gradient-based meta learning framework which explores continuous, dynamic task relationships via task-specific weightings, and can optimise any choice of combination of tasks through the formulation of a meta-loss; where the validation loss automatically influences task weightings throughout training. We apply the proposed framework to both multi-task and auxiliary learning problems in computer vision and robotics, and show that Auto-Lambda achieves state-of-the-art performance, even when compared to optimisation strategies designed specifically for each problem and data domain. Finally, we observe that Auto-Lambda can discover interesting learning behaviors, leading to new insights in multi-task learning. Code is available at https://github.com/lorenmt/auto-lambda.
Code (1)
Tasks
Auxiliary LearningMeta-LearningMulti-Task LearningRobot ManipulationSimilar Papers 제목 키워드 기반
Autonomous Reinforcement of Behavioral Sequences in Neural Dynamics
We introduce a dynamic neural algorithm called Dynamic Neural (DN) SARSA(\lambda) for learning a behavioral sequence from delayed reward. DN-SARSA(\lambda) combines Dynamic Field Theory models of behavioral sequence repr…
General Reinforcement Learningreinforcement-learningReinforcement LearningReinforcement Learning (RL)Adaptive Rate Control for Deep Video Compression with Rate-Distortion Prediction
Deep video compression has made significant progress in recent years, achieving rate-distortion performance that surpasses that of traditional video compression methods. However, rate control schemes tailored for deep vi…
Video CompressionFAVAE: SEQUENCE DISENTANGLEMENT USING IN- FORMATION BOTTLENECK PRINCIPLE
A state-of-the-art generative model, a ”factorized action variational autoencoder (FAVAE),” is presented for learning disentangled and interpretable representations from sequential data via the information bottleneck wit…
DisentanglementRepresentation LearningImplicit Temporal Differences
In reinforcement learning, the TD($\lambda$) algorithm is a fundamental policy evaluation method with an efficient online implementation that is suitable for large-scale problems. One practical drawback of TD($\lambda$) …
Reinforcement LearningSlotFormer: Unsupervised Visual Dynamics Simulation with Object-Centric Models
Understanding dynamics from visual observations is a challenging problem that requires disentangling individual objects from the scene and learning their interactions. While recent object-centric models can successfully …
ObjectQuestion AnsweringVideo PredictionVisual Question Answering+1