Abstract:Multimodal remote sensing data classification can enhance a model's ability to distinguish land features through multimodal data fusion. In this context, how to help models understand the relationship between multimodal data and target tasks has become the focus of researchers. Inspired by the human feedback learning mechanism, causal reasoning mechanism, and knowledge induction mechanism, this paper integrates causal learning, reinforcement learning, and meta learning into a unified remote sensing data classification framework and proposes causal meta-reinforcement learning (CMRL). First, based on the feedback learning mechanism, we overcame the limitations of traditional implicit optimization of fusion features and customized a reinforcement learning environment for multimodal remote sensing data classification tasks. Through feedback interactive learning between agents and the environment, we helped the agents understand the complex relationships between multimodal data and labels, thereby achieving full mining of multimodal complementary information.Second, based on the causal inference mechanism, we designed causal distribution prediction actions, classification rewards, and causal intervention rewards, capturing pure causal factors in multimodal data and preventing false statistical associations between non-causal factors and class labels. Finally, based on the knowledge induction mechanism, we designed a bi-layer optimization mechanism based on meta-learning. By constructing a meta training task and meta validation task simulation model in the generalization scenario of unseen data, we helped the model induce cross-task shared knowledge, thereby improving its generalization ability for unseen multimodal data. The experimental results on multiple sets of multimodal datasets showed that the proposed method achieved state-of-the-art performance in multimodal remote sensing data classification tasks.

Integrating curriculum learning with meta-learning for general rhetoric identification

Class Lifelong Learning for Intent Detection via Structure Consolidation Networks

Ling-CL: Understanding NLP Models through Linguistic Curricula

CERD: A Comprehensive Chinese Rhetoric Dataset for Rhetorical Understanding and Generation in Essays

Curriculum-style Data Augmentation for LLM-based Metaphor Detection

Curriculum Learning for Speech Emotion Recognition from Crowdsourced Labels

Multi-Modal Curriculum Learning over Graphs

Rethinking Meta-Learning from a Learning Lens

MLDGG: Meta-Learning for Domain Generalization on Graphs

LaiDA: Linguistics-aware In-context Learning with Data Augmentation for Metaphor Components Identification

HiCuLR: Hierarchical Curriculum Learning for Rhetorical Role Labeling of Legal Documents

Hacking Task Confounder in Meta-Learning

A Meta Understanding of Meta-Learning

Curriculum Meta-Learning for Next POI Recommendation

Competence-based Curriculum Learning for Multilingual Machine Translation

CITING: Large Language Models Create Curriculum for Instruction Tuning

Causal Meta-Reinforcement Learning for Multimodal Remote Sensing Data Classification

Unsupervised Mutual Learning of Dialogue Discourse Parsing and Topic Segmentation

Clustered Task-Aware Meta-Learning by Learning From Learning Paths

Multi-Task Learning based Online Dialogic Instruction Detection with Pre-trained Language Models

Learning to Program with Natural Language