Learning from Active Human Involvement through Proxy Value Propagation
–Neural Information Processing Systems
Learning from active human involvement enables the human subject to actively intervene and demonstrate to the AI agent during training. The interaction and corrective feedback from human brings safety and AI alignment to the learning process.
Neural Information Processing Systems
Feb-18-2026, 00:01:41 GMT
- Country:
- Oceania > Australia
- New South Wales > Sydney (0.04)
- North America > United States
- Louisiana > Orleans Parish
- New Orleans (0.04)
- California
- San Francisco County > San Francisco (0.14)
- Los Angeles County
- Los Angeles (0.14)
- Long Beach (0.04)
- Louisiana > Orleans Parish
- Europe
- Oceania > Australia
- Genre:
- Research Report > New Finding (0.67)
- Industry:
- Education (1.00)
- Information Technology (0.93)
- Transportation > Ground
- Road (0.67)
- Leisure & Entertainment > Games
- Computer Games (0.93)
- Technology:
- Information Technology > Artificial Intelligence
- Robots (1.00)
- Representation & Reasoning > Agents (1.00)
- Machine Learning
- Reinforcement Learning (1.00)
- Neural Networks (1.00)
- Information Technology > Artificial Intelligence