Abstract:Distance metric learning plays a crucial role in diverse machine learning algorithms and applications. When the labeled information in a target domain is limited, transfer metric learning (TML) helps to learn the metric by leveraging the sufficient information from other related domains. Multitask metric learning (MTML), which can be regarded as a special case of TML, performs transfer across all related domains. Current TML tools usually assume that the same feature representation is exploited for different domains. However, in real-world applications, data may be drawn from heterogeneous domains. Heterogeneous transfer learning approaches can be adopted to remedy this drawback by deriving a metric from the learned transformation across different domains. However, they are often limited in that only two domains can be handled. To appropriately handle multiple domains, we develop a novel heterogeneous MTML (HMTML) framework. In HMTML, the metrics of all different domains are learned together. The transformations derived from the metrics are utilized to induce a common subspace, and the high-order covariance among the predictive structures of these domains is maximized in this subspace. There do exist a few heterogeneous transfer learning approaches that deal with multiple domains, but the high-order statistics (correlation information), which can only be exploited by simultaneously examining all domains, is ignored in these approaches. Compared with them, the proposed HMTML can effectively explore such high-order information, thus obtaining more reliable feature transformations and metrics. Effectiveness of our method is validated by the extensive and intensive experiments on text categorization, scene classification, and social image annotation.

Measuring the Distance Between Finite Markov Decision Processes

A Cantor-Kantorovich Metric Between Markov Decision Processes with Application to Transfer Learning

Measuring and Enhancing the Transferability of Hidden Markov Models for Dynamic Travel Behavioral Analysis

Similarity metrics for Different Market Scenarios in Abides

Deep Model Transferability from Attribution Maps

Measurized Markov Decision Processes Part I: The Discounted Infinite Horizon Criterion

Sufficient Markov Decision Processes with Alternating Deep Neural Networks

Learning Expected Hitting Time Distance

Robust Anytime Learning of Markov Decision Processes

Solving Markov Decision Processes with Reachability Characterization from Mean First Passage Times

On Metrics for Probabilistic Systems: Definitions and Algorithms

A Kernel Perspective on Behavioural Metrics for Markov Decision Processes

Transfer Metric Learning: Algorithms, Applications and Outlooks.

Lipschitz Lifelong Reinforcement Learning

Distance-Dependent Metric Learning.

General Heterogeneous Transfer Distance Metric Learning Via Knowledge Fragments Transfer.

Towards Automatic Measurement Of Probabilistic Processes

Mutual Information Based Knowledge Transfer Under State-Action Dimension Mismatch

Transferring Knowledge Fragments for Learning Distance Metric from A Heterogeneous Domain

Heterogeneous Multitask Metric Learning Across Multiple Domains

Value Function Transfer for Deep Multi-Agent Reinforcement Learning Based on N-Step Returns.