I do research in computer vision, machine learning, and robotics, and I am interested in self-supervised and semi-supervised techniques that allow autonomous agents to learn at low-annotation regimes. My long-term goal is to design learning algorithms that enable autonomous agents to continuously acquire object-centric representations through active perception with multimodal signals for compositional scene understanding and reasoning. My recent effort is to leverage information-theoretic tools to revolutionize scene representations for embodied intelligence.