ARIA Research Lab at NJIT

The ARIA Research Lab in the Department of Computer Science at the Ying Wu College of Computing, New Jersey Institute of Technology (NJIT), led by Dr. Shivvrat Arya, develops methods for trustworthy, structured, and efficient artificial intelligence, integrating learning, reasoning, and optimization to build AI systems that are interpretable, reliable, and scalable. Our research focuses on foundational advances in neurosymbolic AI, probabilistic reasoning, and neural combinatorial optimization, with applications in computer vision, video understanding, and human–AI interaction, including reasoning with large language models.

Research Directions

• Neurosymbolic and Explainable AI We study AI systems that combine neural networks with symbolic logic and probabilistic modeling to encode structure, constraints, prior knowledge, and uncertainty, enabling transparent, controllable reasoning at scale.

• Probabilistic Modeling and Inference We develop tractable and approximate inference methods for generative models, including neural inference engines capable of answering complex queries efficiently in large-scale settings.

• Neural Combinatorial and Constrained Optimization We design learning-based solvers for large-scale discrete and constrained optimization problems, bridging classical combinatorial optimization with modern machine learning.

• Graph Optimization and Structured Decision-Making A major focus is graph-structured optimization, using reinforcement learning and graph neural networks for social network analysis and graph-structured decision-making.

• Applications in Vision, Video, and Human–AI Interaction We apply our methods to real-world domains, including structured and neurosymbolic approaches for video and activity understanding, multimodal learning, augmented reality task guidance, and systems that support effective human–AI collaboration.

Joining the lab

We are always looking for curious, rigorous, and collaborative students who are excited about building the next generation of structured, explainable, and reliable AI systems.

• Ph.D. students: Opportunities to work on core problems in neurosymbolic AI, probabilistic inference, graph optimization, and structured deep learning.
• Undergraduate students: Opportunities to join active projects, learn how to conduct research, and build toward increasingly independent work through options such as CS 488 and CS 489.
• M.S. students: Opportunities for deeper research through faculty-mentored projects or the CS 700B/CS 701B thesis pathway, subject to advisor availability.

If you are interested in joining the ARIA Research Lab, please review the student research opportunities and follow the application instructions there. Briefly describe your background, relevant coursework or projects, and which of the lab’s research directions you are most excited about.

Selected projects & highlights

• NeuPI – Neural Inference Engine
  A neural engine for probabilistic inference that accelerates reasoning in graphical models from minutes to microseconds, enabling real-time decision-making in structured domains.

• CaptainCook4D
  A large-scale egocentric 4D dataset for procedural task understanding, used to study how AI systems perceive and reason about long-horizon activities in realistic, cluttered environments.

• Explainable activity recognition & AR task guidance
  Models that not only recognize what people are doing but also provide structured, interpretable explanations and real-time guidance in augmented reality for complex tasks.

• Award-winning work at top venues
  Lab publications have received best paper awards, spotlights, and oral presentations at venues such as NeurIPS, AISTATS, AAAI, and UAI.