Learning Set Functions that are Sparse in Non-Orthogonal Fourier Bases
نویسندگان
چکیده
Many applications of machine learning on discrete domains, such as preference functions in recommender systems or auctions, can be reduced to estimating a set function that is sparse the Fourier domain. In this work, we present new family algorithms for Fourier-sparse functions. They require at most nk − k log + queries (set evaluations), under mild conditions coefficients, where n size ground and number non-zero coefficients. contrast other work focused orthogonal Walsh-Hadamard transform (WHT), our novel operate with recently introduced non-orthogonal transforms offer different notions Fourier-sparsity. These naturally arise when modeling, e.g., sets items forming substitutes complements. We demonstrate effectiveness several real-world applications.
منابع مشابه
Learning Fourier Sparse Set Functions
Can we learn a sparse graph from observing the value of a few random cuts? This and more general problems can be reduced to the challenge of learning set functions known to have sparse Fourier support contained in some collection P. We prove that if we choose O(k log |P|) sets uniformly at random, then with high probability, observing any k-sparse function on those sets is sufficient to recover...
متن کاملBases and Transforms of Set Functions
The paper studies the vector space of set functions on a finite set X, which can be alternatively seen as pseudo-Boolean functions, and including as a special cases games. We present several bases (unanimity games, Walsh and parity functions) and make an emphasis on the Fourier transform. Then we establish the basic duality between bases and invertible linear transform (e.g., the Möbius transfo...
متن کاملLearning Sparse Polynomial Functions
We study the question of learning a sparse multivariate polynomial over the real domain. In particular, for some unknown polynomial f(~x) of degree-d and k monomials, we show how to reconstruct f , within error , given only a set of examples x̄i drawn uniformly from the n-dimensional cube (or an n-dimensional Gaussian distribution), together with evaluations f(x̄i) on them. The result holds even ...
متن کاملQuasi-orthogonal expansions for functions in BMO
For {φ_n(x)}, x ε [0,1] an orthonormalsystem of uniformly bounded functions, ||φ_n||_{∞}≤ M
متن کاملLearning Heterogeneous Functions from Sparse and Non-Uniform Sample
A boosting-based method for centers placement in radial basis function networks (RBFN) is proposed. Also, the influence of several methods for drawing random samples on the accuracy of RBFN is examined. The new method is compared to trivial, linear and non-linear regressors including the multilayer Perceptron and alternative RBFN learning algorithms and its advantages are demonstrated for learn...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Proceedings of the ... AAAI Conference on Artificial Intelligence
سال: 2021
ISSN: ['2159-5399', '2374-3468']
DOI: https://doi.org/10.1609/aaai.v35i12.17232