Postdoctoral Associate

Project Overview We are seeking one Postdoctoral Associate in Computer Science to join the "Towards an Immersive Workplace" project. This project is part of Team "T1" of a five-year, ambitious research program on "Mens, Manus and Machina-How AI Empowers People, Institutions and Cities in Singapore (M3S)."Successful applicants for this position will have the opportunity to work on cutting-edge research that builds breakthrough new capabilities for AI-enhanced human-robot collaborative execution of tasks that involve sophisticated spatial reasoning over unstructured, physical environments (such as homes, offices, industrial sites and shopping malls) and alignment of such spatial context with procedural models of task execution. There will be a special focus on optimizing the computational load of such AI models to support their execution on resource-constrained pervasive and edge devices. The end goal is to build exciting new capabilities that allow robots and humans to work together, interactively and in close proximity, such that the robotic actions are controlled both by (i) commands and queries, often at higher-levels of abstraction, that humans would issue naturally in untethered, field settings and (ii) explicit task execution knowledge embedded in multi-modal documents.This research will specifically look at the development of embodied and embedded intelligence, which allows AI-based agents embedded in robots to perform sophisticated spatial reasoning to formulate a sequence of manipulation tasks that help achieve various human-specified objectives. This will require development of new AI and systems capabilities along multiple dimensions, including: (a) developing a deep semantic understanding of 3D physical spaces, including the location and relationship of various objects located within such spaces; (b) translating higher-order, qualitative human commands and task descriptions into a set of physically-grounded, objective tasks; and (c) further adapting such task execution, including manipulation actions, in response to the evolving activity of human co-workersIn addition, the research team will also address the challenge of runtime optimization of the associated AI models such that they are amenable to low-latency, interactive execution on embedded devices.The post-doctoral researcher will be part of an extended team, comprising additional post-doctoral scientists and research engineers who are exploring cognate research areas, including (a) improved visual grounding of natural multi-modal (combining visual, verbal and gestural cues) human queries and instructions, (b) innovations in multi-modal sensing (e.g., fusion of event camera, LIDAR and RGB vision sensing streams) that permit ultra-low power, saliency-driven sensing of 3D physical spaces, and (c) unobtrusive capture and use of implicit human attention and expertise for natural human-machine collaboration. The team is also working to demonstrate the impact of such AI-driven advances via selected, real-world pilot technology demonstration projects with leading industry collaborators.The T1 team is led by distinguished scholars, including Profs. Sanjay Sarma, Daniela Rus and Jinhua Zhao from MIT; and Prof. Archan Misra from Singapore Management University. In Singapore, proposed research described above will be coordinated by Prof. Misra.Key Responsibilities Conduct and coordinate research on the design, development and testing of new AI models and techniques for embedded execution of embodied AI and spatial intelligence. Publish high quality research results in top-tier conferences and journals, with focus on venues associated with AI, robotic intelligence and mobile/embedded systems (e.g., AAAI, NeurIPS, ACM Multimedia, ICCV, IEEE IROS, IEEE ICRA ) . Participate in the mentorship and training of graduate and undergraduate students at MIT and in Singapore, and research engineers based in Singapore Assist in grant writing, project management, and other administrative duties related to research activities. Requirements Ph.D. in Computer Science, on topics related to Artificial Intelligence and Machine Learning, Human-Robot Interaction and 3D Vision/Multimedia. Experience with deep neural models and algorithms for 3D spatial reasoning and/or robotic planning Experience working with multi-modal AI models combining language, vision, gestures and gaze modalities. Strong publication record in top-tier conferences and journals, especially in the areas of AI and/or Human-Robot Collaboration (e.g., AAAI, ACM Multimedia, ICCV, IEEE IROS, IEEE ICRA) . DESIRABLE: Demonstrated ability to conduct inter-disciplinary research with a systems focus. Excellent communication and collaboration skills. To find out more about this role, please contact Professor Archan Misra ( ).To apply, please visit our website at: applicants are invited to send in their full CV/resume, cover letter and list of three references (to include reference names and contact information). We regret that only shortlisted candidates will be notified.