About Me

I am a second year Ph.D. student in the College of Engineering and Computer Science (CECS) at the Australian National University, where I am fortunate to be supervised by Prof. Liang Zheng and Prof. Stephen Gould. I am also an intern in the Creative Intelligence Lab at Adobe Research working under Dr. Zhe Lin and Dr. Jianming Zhang.

Ph.D. Research Topic: Controllable Image Synthesis, Creative Content Generation.

Previously, I was a research intern at Adobe Research. Prior to that, I had graduated from Masters of Machine learning and Computer Vision with first class honours from the Australian National University. Even earlier, I was working as a machine learning research engineer at Yahoo Japan. I received my Bachelors in Electrical Engineering with specialization in intelligent and cognitive systems from IIT Delhi.

Publications

	High-Fidelity Guided Image Synthesis with Latent Diffusion Models Jaskirat Singh, Stephen Gould, and Liang Zheng. CVPR 2023 [Paper] [Code] [Project Website]
	Paint2Pix: Interactive Painting based Progressive Image Synthesis and Editing Jaskirat Singh, Liang Zheng, Cameron Smith, and Jose Echevarria. ECCV 2022 [Paper] [Code] [Project Website]
	Intelli-Paint: Towards Developing More Human-Intelligible Painting Agents Jaskirat Singh, Cameron Smith, Jose Echevarria, and Liang Zheng. ECCV 2022, US Patent [Paper] [Project Website]
	Combining Semantic Guidance and Deep Reinforcement Learning for Generating Human Level Paintings Jaskirat Singh and Liang Zheng, 2021. CVPR 2021 [Paper] [Code]
	Enhanced Scene Specificity with Sparse Dynamic Value Estimation Jaskirat Singh and Liang Zheng, 2020 arXiv 2021 [Paper]
	Dynamic Value Estimation for Single-Task Multi-Scene Reinforcement Learning Jaskirat Singh and Liang Zheng, 2020 arXiv 2021 [Paper]

	Navigation Trained a Deep Reinforcement Learning Agent to navigate a world simulated in the Unity Environment. The underlying model was a Dueling Double Deep Q Network (DDQN) with prioritized experience replay.
	Flying a Quadcopter Trained an Reinforcement learning based agent to learn how to fly a quadcopter Actor-Critic based DDPG algorithm is used along with prioritized experience replay
	Robotic arm control Trained mutliple Double-jointed robotic arms to reach target locations using PPO algorithm Uses multiple (non-interacting, parallel) copies of the same agent to distribute the task of gathering experience
	Multi agent Collaboration/Competition Trained a pair of RL agents to play tennis using Multi Agent DDPG (MADDPG) algorithm The model consists of a centralized critic for each agent with access to combined state and action values from all agents
	Alphazero for Tictactoe Alphazero implementation for the game of Tictactoe Extended the solution to a larger and difficult 6-6-4 Tictactoe.
	Memory Matching Game Interface Implementation of memory matching game using javascript and DOM
	Arcade Game Classic Arcade Game built using Object oriented Design patterns