Seminar Series - Scalable Learning Adaptive to Unknown Dynamics and Graphs. Professor Yanning Shen, EECS

Tuesday, May 26 at 2 p.m. via Zoom


Abstract: We live in an era of data deluge, where pervasive media collect massive amounts of data, often in a streaming fashion. A major portion of the data resides on networks representing a wide range of physical, biological, social, and financial interdependencies, e.g., brain networks, gene regulatory networks, and smart grids. Learning from these dynamic and large volumes of (network) data is hence expected to bring significant science and engineering advances along with consequent improvements in quality of life. However, with the blessings come big challenges. The sheer volume of data makes it impossible to run analytics in batch form. Large-scale datasets are noisy, incomplete, and prone to outliers. As many sources continuously generate data in real time, it is often impossible to store all of it. Thus, analytics must often be performed in real time, without a chance to revisit past entries. Furthermore, the networks on which the data reside can have very large size, and nodal attributes can be unavailable, e.g., due to privacy concerns. Meanwhile, new nodes can emerge over time, which can necessitate real-time evaluation of their nodal attributes. In response to these challenges, this talk will first introduce an online scalable function approximation scheme that is suitable for various machine learning tasks. The novel approach adaptively learns and tracks the sought nonlinear function `on the fly' with quantifiable performance guarantees, even in adversarial environments with unknown dynamics. Building on this robust and scalable function approximation framework, a privacy-preserving graph-aware learning approach will be outlined next for learning over time-varying graphs with possibly growing sizes. Effectiveness of the novel algorithms will be showcased in several real-world datasets.