Search for tag: "triad distinguished lecture series"
Yuxin Chen - Inference and Uncertainty Quantification for Noise Matrix CompletionNoisy matrix completion aims at estimating a low-rank matrix given only partial and corrupted entries. Despite substantial progress in designing efficient estimation algorithms, it remains largely…
From Katie Gentilello
51 plays
0
|
|
Yuxin Chen - Random initialization and implicit regularization in nonconvex statistical estimationRecent years have seen a flurry of activities in designing provably efficient nonconvex procedures for solving statistical estimation / learning problems. Due to the highly nonconvex nature of the…
From Katie Gentilello
26 plays
0
|
|
Yuxin Chen - Spectral Methods Meets Asymmetry: Two Recent StoriesThis talk is concerned with the interplay between asymmetry and spectral methods. Imagine that we have access to an asymmetrically perturbed low-rank data matrix. We attempt estimation of the…
From Katie Gentilello
20 plays
0
|
|
Yuxin Chen - Projected Power Method: An Efficient Algorithm for Joint Discrete AssignmentVarious applications involve assigning discrete label values to a collection of objects based on some pairwise noisy data. Due to the discrete---and hence nonconvex---structure of the problem,…
From Katie Gentilello
27 plays
0
|
|
Yuxin Chen - [Lecture 1] The power of nonconvex optimization in solving random quadratic systems of equationsWe consider the fundamental problem of solving random quadratic systems of equations in n variables, which spans many applications ranging from the century-old phase retrieval problem to various…
From Katie Gentilello
62 plays
0
|
|
Johannes Schmidt-Hieber - Mathematics for Deep Neural Networks: Energy landscape and open problems (Lecture 5/5)To derive a theory for gradient descent methods, it is important to have some understanding of the energy landscape. In this lecture, an overview of existing results is given. The second part of the…
From Katie Gentilello
90 plays
0
|
|
Johannes Schmidt-Hieber - Mathematics for Deep Neural Networks: Statistical theory for deep ReLU networks (Lecture 4/5)We outline the theory underlying the recent bounds on the estimation risk of deep ReLU networks. In the lecture, we discuss specific properties of the ReLU activation function that relate to skipping…
From Katie Gentilello
88 plays
0
|
|
Johannes Schmidt-Hieber - Mathematics for Deep Neural Networks: Advantages of Additional Layers (Lecture 3/5)Why are deep networks better than shallow networks? We provide a survey of the existing ideas in the literature. In particular, we discuss localization of deep networks, functions that can be easily…
From Katie Gentilello
94 plays
0
|
|
Johannes Schmidt-Hieber - Mathematics for Deep Neural Networks : Theory for shallow networks (Lecture 2/5)We start with the universal approximation theorem and discuss several proof strategies that provide some insights into functions that can be easily approximated by shallow networks. Based on this, a…
From Katie Gentilello
156 plays
0
|
|
Johannes Schmidt-Hieber - Mathematics for Deep Neural Networks (Lecture 1/5)Lecture 1) Survey on neural network structures and deep learning There are many different types of neural networks that differ in complexity and the data types that can be processed. This lecture…
From Katie Gentilello
275 plays
0
|
|
Michael Mitzenmacher - Bloom Filters, Cuckoo Hashing, Cuckoo Filters, Adaptive Cuckoo Filters, and Learned Bloom FiltersI will go over some of my past and present work on hashing-based data structures. After presenting some background on Bloom filters and cuckoo hashing, we will describe cuckoo filters, an efficient…
From Katie Gentilello
342 plays
0
|
|
Gabor Lugosi - Mean Estimation: Median-of-Means TournamentsIn these lectures we discuss some statistical problems with an interesting combinatorial structure behind. We start by reviewing the "hidden clique" problem, a simple prototypical example…
From Katie Gentilello
32 plays
0
|
|
Gabor Lugosi - Combinatorial Testing ProblemsIn these lectures we discuss some statistical problems with an interesting combinatorial structure behind. We start by reviewing the "hidden clique" problem, a simple prototypical example…
From Katie Gentilello
32 plays
0
|
|
Sara van de Geer - The Debiased LassoIn the third lecture we use sparsity to establish confidence intervals for a parameter of interest. The idea is to use the penalized estimator as an initial estimator in a one-step Newton-Raphson…
From Katie Gentilello
254 plays
0
|
|
Sara van de Geer - Compatibility and the LassoWe will see in the first lecture that one needs conditions which relate the penalty to the risk function. They have in a certain sense to be “compatible”. We discuss these compatibility…
From Katie Gentilello
152 plays
0
|