So sánh phương pháp
Xem các phương pháp đã chọn cạnh nhau; những hàng khác biệt được làm nổi bật.
| Học tăng cường đa phương thức× | Mạng nơ-ron đồ thị đa phương thức× | |
|---|---|---|
| Lĩnh vực | Học sâu | Học sâu |
| Họ | Machine learning | Machine learning |
| Năm ra đời≠ | 2015–2022 | 2019–2020 |
| Người khởi xướng≠ | Multiple contributors (DeepMind, OpenAI, Google Brain, 2010s–2020s) | Kipf & Welling (GNN foundation); extended to multimodal settings by multiple research groups c. 2019–2020 |
| Loại≠ | Multimodal deep RL agent | Graph-based deep learning with multimodal input fusion |
| Công trình gốc≠ | Reed, S., Zolna, K., Parisotto, E., Colmenarejo, S. G., Novikov, A., Barth-Maron, G., ... & de Freitas, N. (2022). A Generalist Agent. Transactions on Machine Learning Research. link ↗ | Kipf, T. N., & Welling, M. (2017). Semi-Supervised Classification with Graph Convolutional Networks. International Conference on Learning Representations (ICLR). link ↗ |
| Tên gọi khác | Multimodal RL, Multi-Sensory Reinforcement Learning, Vision-Language RL, Multi-Input RL | MM-GNN, Multimodal GNN, Multi-modal Graph Network, Cross-modal Graph Neural Network |
| Liên quan | 6 | 6 |
| Tóm tắt≠ | Multimodal Reinforcement Learning trains agents to make sequential decisions by perceiving and integrating multiple input modalities — such as raw pixels, language instructions, audio, and proprioceptive sensors — simultaneously. Rather than acting on a single data stream, the agent fuses heterogeneous signals into a unified state representation and learns a policy through environmental reward feedback. | A Multimodal Graph Neural Network (MM-GNN) combines data from multiple modalities — such as text, images, and structured features — into a unified graph structure and applies graph-based message passing to learn joint representations. It enables relational reasoning across heterogeneous data sources, going beyond what unimodal or simple concatenation approaches can capture. |
| ScholarGateBộ dữ liệu ↗ |
|
|