Akanksha Saran
Towards Real-World Deployment of Interactive Machine Learning
Research Abstract:
Systems that can learn interactively from their end-users are quickly becoming widespread in real-world applications. Traditional interactive learning frameworks often expect human input to be provided in the form of labels, scalar feedback, or demonstrations. However, for many interactive learning problems in the real-world, it is not possible to provide hand-defined labels for each data point. Moreover, hand-defined rewards require domain expertise and are cumbersome to design. On the other hand, demonstrations are more intuitive to provide, but do not fully capture expressive signals of human intent and are also expensive to collect. These challenges necessitate the design of machine learning algorithms which efficiently leverage human input in more diverse forms such as multimodal interactive feedback, active queries, and preferences. My research introduces several interactive learning algorithms which reduce the labeling burden, leverage diverse modalities of human feedback, and alleviate the need for hand-defined reward functions.
Bio:
Akanksha is a postdoctoral researcher at Microsoft Research NYC. Her research focuses on modeling human behavior across multimodal interfaces and designing human-interactive machine learning algorithms for sequential decision-making settings such as imitation learning, interaction-grounded learning, and active learning. She is particularly interested in alleviating barriers of access for end-users who can benefit from interactive machine learning algorithms in the real-world. Her interdisciplinary work attempts to (1) further our understanding of how humans interact with artificial learning agents and (2) design efficient algorithms which leverage varied forms of human data for real-world interactive applications. Akanksha obtained her PhD in Computer Science from UT Austin, MS in Robotics from CMU, and Bachelors in Computer Science and Engineering from IIT Jodhpur in India. She is the recipient of the Google Anita Borg Memorial Award, NSF Doctoral Consortium Award, President's Gold Medal at IIT Jodhpur, and recognized as a Pioneer in the field of Human-Robot Interaction.