Daison Darlan
Papers
2
Total Citations
4
H-Index
2
About
Daison Darlan is a researcher at the forefront of human-robot interaction and reinforcement learning, with a focus on making machines more socially and emotionally intelligent. His work bridges affective computing and autonomous decision-making, addressing how robots can interpret human intent through touch and optimize complex behaviors. In his 2023 study, "Recognizing Social Touch Gestures using Optimized Class-weighted CNN-LSTM Networks," Darlan developed a deep learning framework that significantly improves the recognition of affective touch gestures—a critical capability for companion and therapeutic robots. By employing a class-weighted CNN-LSTM architecture, his model enhances accuracy in distinguishing subtle emotional cues conveyed through physical contact, earning early citations for its practical relevance. Expanding into multi-objective reinforcement learning, his 2024 paper, "Prediction-guided multi-objective reinforcement learning with corner solution search," introduces a novel algorithm that efficiently balances competing goals in robotic tasks, offering a scalable approach for real-world applications. With these contributions, Darlan is shaping the future of socially aware robotics, where machines not only perceive human emotions but also make intelligent, adaptive decisions.
Research Focus
Key Achievements
Top Papers
- 1
- 2