We describe a primal-dual framework for the design and analysis of online convex optimization algorithms for drifting regret. Existing literature shows (nearly) optimal drifting regret bounds only for the l2 and the l1-norms. Our work provides a connection between these algorithms and the Online Mirror Descent (OMD) updates; one key insight that results from our work is that in order for these ...

Our group worked on accelerating the X-ray Computed Tomography (CT) reconstruction with statistical image reconstruction algorithm, using single-instruction multiple data (SIMD) instructions to accelerate the regularizer part. Also, we used half-width floating point data format to mitigate the memory bandwidth problem. Our results show that we could achieve 2.5x speedup by combining SIMD instru...

A popular approach for estimating an unknown signal x0 ∈ R from noisy, linear measurements y = Ax0 +z ∈ R is via solving a so called regularized M-estimator: x̂ := arg minx L(y−Ax)+λf(x). Here, L is a convex loss function, f is a convex (typically, non-smooth) regularizer, and, λ > 0 is a regularizer parameter. We analyze the squared error performance ‖x̂ − x0‖2 of such estimators in the high-dim...

In this work we introduce a new approach to co-classification, i.e. the task of jointly classifying multiple, otherwise independent, data samples. The method we present, named CoConut, is based on the idea of adding a regularizer in the label space to encode certain priors on the resulting labelings. A regularizer that encourages labelings that are smooth across the test set, for instance, can ...