Dimensions overall score 7.0

GitHub Code Pulse

No public code linked for this paper yet.

• The proposed negative-feedback system achieves a 90% improvement in success rate against a system that does not utilize negative feedback for ambiguous tasks in simulation.result 95%
• The proposed negative-feedback system achieves a 50% improvement in success rate when utilized on a real robot for ambiguous tasks.result 95%
• The novel negative learning algorithm is based on product of experts methods.method 90%
• The product of experts method allows negative feedback to have maximum impact on an undesired region of the state space.technical 90%
• Learning only the negative policy at each feedback step minimizes learning time.technical 90%
• The system is capable of learning from suboptimal demonstrations.technical 85%
• The novel scheme is validated through simulated and real-robot experiments.method 95%
• The proposed system demonstrates higher efficacy, memory efficiency, and time efficiency than a comparable negative feedback scheme.result 95%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card