A global convergence theory for deep ReLU implicit networks via over-parameterization

Kavli Affiliate: Jia Liu | First 5 Authors: Tianxiang Gao, Hailiang Liu, Jia Liu, Hridesh Rajan, Hongyang Gao | Summary: Implicit deep learning has received increasing attention recently due to the fact that it generalizes the recursive prediction rules of many commonly used neural network architectures. Its prediction rule is provided implicitly based on the […]


Continue.. A global convergence theory for deep ReLU implicit networks via over-parameterization

Quick-Look Pipeline Lightcurves for 9.1 Million Stars Observed Over the First Year of the TESS Extended Mission

Kavli Affiliate: George Ricker | First 5 Authors: Michelle Kunimoto, Chelsea Huang, Evan Tey, Willie Fong, Katharine Hesse | Summary: We present a magnitude-limited set of lightcurves for stars observed over the TESS Extended Mission, as extracted from full-frame images (FFIs) by MIT’s Quick-Look Pipeline (QLP). QLP uses multi-aperture photometry to produce lightcurves for ~1 […]


Continue.. Quick-Look Pipeline Lightcurves for 9.1 Million Stars Observed Over the First Year of the TESS Extended Mission

Graph-Guided Network for Irregularly Sampled Multivariate Time Series

Kavli Affiliate: Xiang Zhang | First 5 Authors: Xiang Zhang, Marko Zeman, Theodoros Tsiligkaridis, Marinka Zitnik, | Summary: In many domains, including healthcare, biology, and climate science, time series are irregularly sampled with varying time intervals between successive readouts and different subsets of variables (sensors) observed at different time points. Here, we introduce RAINDROP, a […]


Continue.. Graph-Guided Network for Irregularly Sampled Multivariate Time Series

Online Graph Learning in Dynamic Environments

Kavli Affiliate: Xiang Zhang | First 5 Authors: Xiang Zhang, , , , | Summary: Inferring the underlying graph topology that characterizes structured data is pivotal to many graph-based models when pre-defined graphs are not available. This paper focuses on learning graphs in the case of sequential data in dynamic environments. For sequential data, we […]


Continue.. Online Graph Learning in Dynamic Environments

Time-varying Graph Learning Under Structured Temporal Priors

Kavli Affiliate: Xiang Zhang | First 5 Authors: Xiang Zhang, Qiao Wang, , , | Summary: This paper endeavors to learn time-varying graphs by using structured temporal priors that assume underlying relations between arbitrary two graphs in the graph sequence. Different from many existing chain structure based methods in which the priors like temporal homogeneity […]


Continue.. Time-varying Graph Learning Under Structured Temporal Priors