Algorithms above the noise floor
S Ludwig - 2018 - dspace.mit.edu
Many success stories in the data sciences share an intriguing computational phenomenon.
While the core algorithmic problems might seem intractable at first, simple heuristics or …
While the core algorithmic problems might seem intractable at first, simple heuristics or …
Approximate optimization of convex functions with outlier noise
We study the problem of minimizing a convex function given by a zeroth order oracle that is
possibly corrupted by {\em outlier noise}. Specifically, we assume the function values at …
possibly corrupted by {\em outlier noise}. Specifically, we assume the function values at …
Nearly tight bounds for discrete search under outlier noise
Binary search is one of the most fundamental search routines, exploiting the hidden
structure of the search space. In particular, it cuts down exponentially on the complexity of …
structure of the search space. In particular, it cuts down exponentially on the complexity of …
Statistical optimization in high dimensions
We consider optimization problems whose parameters are known only approximately,
based on a noisy sample. Of particular interest is the high-dimensional regime, where the …
based on a noisy sample. Of particular interest is the high-dimensional regime, where the …
Noise Stability Optimization For Flat Minima With Tight Rates
Generalization properties are a central aspect of the design and analysis of learning
algorithms. One notion that has been considered in many previous works as leading to good …
algorithms. One notion that has been considered in many previous works as leading to good …
[图书][B] Statistical inference via convex optimization
A Juditsky, A Nemirovski - 2020 - books.google.com
This authoritative book draws on the latest research to explore the interplay of high-
dimensional statistics with optimization. Through an accessible analysis of fundamental …
dimensional statistics with optimization. Through an accessible analysis of fundamental …
Algorithm portfolios for noisy optimization: Compare solvers early
Noisy optimization is the optimization of objective functions corrupted by noise. A portfolio of
algorithms is a set of algorithms equipped with an algorithm selection tool for distributing the …
algorithms is a set of algorithms equipped with an algorithm selection tool for distributing the …
Algorithms and matching lower bounds for approximately-convex optimization
A Risteski, Y Li - Advances in Neural Information Processing …, 2016 - proceedings.neurips.cc
In recent years, a rapidly increasing number of applications in practice requires solving non-
convex objectives, like training neural networks, learning graphical models, maximum …
convex objectives, like training neural networks, learning graphical models, maximum …
Extreme points under random noise
V Damerow, C Sohler - European Symposium on Algorithms, 2004 - Springer
Given a point set P={p_1,\dots,p_n\} in the d-dimensional unit hypercube, we give upper
bounds on the maximal expected number of extreme points when each point pi is perturbed …
bounds on the maximal expected number of extreme points when each point pi is perturbed …
On multiplicative noise models for stochastic search
M Jebalia, A Auger - International Conference on Parallel Problem Solving …, 2008 - Springer
In this paper we investigate multiplicative noise models in the context of continuous
optimization. We illustrate how some intrinsic properties of the noise model imply the failure …
optimization. We illustrate how some intrinsic properties of the noise model imply the failure …