Dr. Ajaz Ahmad Bhat

ajaz.bhat@ubd.edu.bn

Program Leader, School of Digital Science and Assistant Professor, School of Digital Science



                    

Dr Ajaz Ahmad Bhat is an Assistant Professor of Intelligent Robotics at the Faculty of Science, Universiti Brunei Darussalam, Brunei, and a visiting researcher at the Okinawa Institute of Science and Technology, Japan. Prior to this posting, he worked as a Senior Research Associate at the University of East Anglia, UK, and then as an Assistant Professor at the University of Twente, Netherlands. Ajaz’s research interests lie at the intersection of intelligent robotics, cognitive sciences, and consumer technologies with a special focus on understanding and designing developmental and cumulative learning systems. His research outputs have been published in more than 15 journal articles with some of them in top-tier journals such as Psychological Review and Autonomous Robots. His work on Causal Learning in Robots was covered in top conferences like ICML and by media outlets such as BBC. He has been involved in multiple international research projects such as EU’s FP7 DARWIN and NIH’s WOLVES. Ajaz retains memberships of both IEEE and APS societies in addition to holding a Ph.D. in Cognitive Robotics from the Istituto Italiano di Tecnologia (IIT), Italy.



************************
Information for PhD/Masters Aspirants
************************
Vacancies are available for local and international candidates interested in pursuing a Master's or PhD degree in our lab. Our major research areas include Artificial General Intelligence, Cognitive Robotics, Natural Language Understanding, Neuro-cognitive Development, and Advanced Machine Learning. Some of the projects our lab is pursuing are listed below that may interest you. To apply to our lab, please email me your research proposal, CV, academic transcripts, and any publications, demonstrating your technical knowledge in any/all of the areas above.

The following links will guide you to the admission/application process
Postgraduate admissions: https://ubd.edu.bn/admission/graduate/
Scholarships available for eligible applicants: https://ubd.edu.bn/admission/scholarship/
Other Scholarships: https://www.mfa.gov.bn/Pages/bdgs/bdgs2022.aspx

The admission intake occurs twice a year as follows:
August Intake: (Application time: Around January)
January Intake: (Application time: Around July)

EDUCATION

PhD in Cognitive Robotics & AI
Robotics, Brain and Cognitive Sciences Department,
Italian Institute of Technology,
Genova, Italy
(Jan 2013- Apr 2016)


Masters in Computer Applications (Professional),
Department of Computer Science,
University of Kashmir, Srinagar, India
(Mar 2007 - Jun 2010)

Bachelors in Physical Sciences,
University of Kashmir, India
(Mar 2004- Feb 2007)

RESEARCH INTERESTS

• Connectionist, dynamical, multi-time scale, and integrated systems approach to intelligence
• AI robots for problem-solving, motor behaviours, causal reasoning, imagination, and creativity
• Neuro-symbolic models of episodic, semantic, procedural, working memories and selective attention
• Cognition-inspired intelligent systems, human neuro-cognition, and cognitive development
• Grounded language learning, attention-guided neural networks, and explainable AI

FUTURE PROJECTS

Cognitive Continual Learning Systems for Reasoning and Artificial General Intelligence (AGI)

Deep learning excels at training powerful models from fixed datasets and stationary environments, often exceeding human-level ability. Yet, these models fail to emulate human learning, which is robust, incremental, compositional, constructive, and predictive from sequential experience to reason beyond experience. This project investigates mechanisms that can encode, recall and exploit diverse past experiences of an agent to imagine the future and solve non-trivial problems creatively. Approaches include at the most granular level, with gradient-based methods, as well as at the architectural level, with modular and memory-based, and meta-learning methods. The project will help understand the theoretical capabilities and limitations of AI, it will also pave the way towards creating AI systems that are explainable, controllable, reliable, and dependable — a major challenge with existing AI. Methodologies include dynamical systems, neuro-symbolic AI, episodic memory, predictive coding, reinforcement, self-supervising, and self-organizing networks.


Applications Invited
For PhD/MS Students

Compositional Action and Motion Planning in Robots

How do humans imitate, learn and recycle skilled actions including the use of tools? This project develops different representations of the body, movement, and peri-personal spaces to enable learning of dynamic task-specific trajectory planning, body-chain engagement, motor control, and prediction. The project investigates how the motor knowledge acquired by a robot during learning a skill like drawing can be recycled in a completely different skill like the use of a tool. The project will look for designing a full-blown procedural memory and identifying mechanisms that allow humans (and robots) to generate any form of complex actions at runtime through shape compositionality and recycling of stable/metastable movement patterns emergent at phase transitions in different neural states. Methodologies employed include neuro-symbolic AI, predictive coding, control theory, RL, deep learning transformers, and self-organizing networks.


Applications Invited
For PhD/MS Students

Grounded Language Learning and Understanding beyond NLP

Language is grounded in sensory-motor experiences not word embeddings. Agents must interact physically with their world to grasp the essence of words and context. This project builds agents (and interactive robots) in simulated environments that learn and understand language via multisensory grounding (and robotic embodiment). The goals are two-fold. First, the project is using computational models to gain insights into how children acquire language over development. By building and testing models with near-realistic environments, we test psychological theories of human cognition and language acquisition. Second, we build a new generation of instructible language models with richer semantic representations leading to more intelligent machine behavior. Methodologies employed include dynamic field theory, neuro-symbolic AI, deep reinforcement learning, and transformer networks.


Applications Invited
For PhD/MS Students

RECENT PUBLICATIONS

Bhat A. A., J Spencer, L Samuelson (2021). Word-Object Learning via Visual Exploration in Space (WOLVES): A Neural Process Account of Cross-Situational Word Learning. https://doi.org/10.1037/rev0000313, Psychological Review

Bhat A. A., Mohan V. (2020). Causal Learning by a Robot with Semantic-Episodic Memory in an Aesop's Fable Experiment, International Conference on Learning and Representation, (ICLR), Addis Ababa arXiv:2003.00274 https://arxiv.org/abs/2003.00274

Mohan V., Bhat A. A., Morasso P. (2019). Muscle-less motor synergies and actions without movements. From motor neuroscience to cognitive robotics. Physics of Life Reviews. https://doi.org/10.1016/j.plrev.2018.04.005

Bhat A. A., Mohan V. (2018). Goal directed reasoning and cooperation in robots in shared workspaces: An internal simulation based neural framework, Cognitive Computation https://doi.org/10.1007/s12559-018-9553-1

Mohan, V., & Bhat, A. A. (2018). Joint Goal Human-Robot collaboration-From Remembering to Inferring. 8th Annual International Conference on Biologically Inspired Cognitive Architectures BICA 2017, published in Procedia Computer Science, 123, 579-584. [Full paper] https://doi.org/10.1016/j.procs.2018.01.089