Research
The following is a very brief description of my active research topics. Please send me an email for more details on active projects
Generative AI: Exploring cutting-edge techniques like Variational Autoencoders (VAEs) and Diffusion Models to generate new, realistic data samples from learned distributions. These methods are pivotal for advancing synthetic data generation, enhancing privacy, and improving data diversity.
Adversarial Purification: Utilizing generative AI models to cleanse data inputs from potential adversarial attacks, thereby safeguarding AI systems against manipulation and enhancing model robustness.
Reinforcement Learning (RL): Focusing on the development and refinement of RL techniques, including:
Deep Reinforcement Learning: Leveraging deep neural networks to enable agents to learn optimal strategies in complex, high-dimensional environments.
Hierarchical Reinforcement Learning: Breaking down tasks into simpler sub-tasks to streamline the learning process and improve decision-making efficiency.
Integrating Generative Models and RL: Innovating at the intersection of generative models and reinforcement learning to create more adaptable, efficient, and robust AI systems. This includes generating dynamic training environments and simulating future scenarios to enhance decision-making processes.