Budhitama Subagdja

Papers

Total Citations

H-Index

About

Budhitama Subagdja is a researcher at the intersection of artificial intelligence, multi-agent systems, and cognitive robotics. His work focuses on enabling autonomous agents—from robots to conversational AI—to collaborate, learn, and even exhibit self-awareness. In his most cited work, "End-to-end Deep Reinforcement Learning for Multi-agent Collaborative Exploration" (2019, 25 citations), Subagdja tackles a core challenge in robotics: coordinating multiple agents to explore unknown environments without interference. He introduces a CNN-based deep reinforcement learning model that allows agents to learn collaborative strategies end-to-end, significantly improving efficiency in tasks like search and rescue. Earlier, his "Interactive Teachable Cognitive Agents" framework (2016, 10 citations) proposed a modular approach to building complex multi-agent systems, where smart components can be taught and reused—reducing development time for applications requiring distributed intelligence. Subagdja also explores the frontier of socially aware AI. In "Towards a Brain Inspired Model of Self-Awareness for Sociable Agents" (2017, 4 citations), he draws on cognitive neuroscience to design agents that can reflect on their own thoughts and experiences, enabling more natural human-robot interaction. His work bridges practical engineering with foundational questions about machine consciousness, making him a distinctive voice in the quest for truly intelligent, collaborative agents.