English

Memory-Sample Lower Bounds for Learning Parity with Noise

Machine Learning 2021-07-07 v1 Computational Complexity

Abstract

In this work, we show, for the well-studied problem of learning parity under noise, where a learner tries to learn x=(x1,,xn){0,1}nx=(x_1,\ldots,x_n) \in \{0,1\}^n from a stream of random linear equations over F2\mathrm{F}_2 that are correct with probability 12+ε\frac{1}{2}+\varepsilon and flipped with probability 12ε\frac{1}{2}-\varepsilon, that any learning algorithm requires either a memory of size Ω(n2/ε)\Omega(n^2/\varepsilon) or an exponential number of samples. In fact, we study memory-sample lower bounds for a large class of learning problems, as characterized by [GRT'18], when the samples are noisy. A matrix M:A×X{1,1}M: A \times X \rightarrow \{-1,1\} corresponds to the following learning problem with error parameter ε\varepsilon: an unknown element xXx \in X is chosen uniformly at random. A learner tries to learn xx from a stream of samples, (a1,b1),(a2,b2)(a_1, b_1), (a_2, b_2) \ldots, where for every ii, aiAa_i \in A is chosen uniformly at random and bi=M(ai,x)b_i = M(a_i,x) with probability 1/2+ε1/2+\varepsilon and bi=M(ai,x)b_i = -M(a_i,x) with probability 1/2ε1/2-\varepsilon (0<ε<120<\varepsilon< \frac{1}{2}). Assume that k,,rk,\ell, r are such that any submatrix of MM of at least 2kA2^{-k} \cdot |A| rows and at least 2X2^{-\ell} \cdot |X| columns, has a bias of at most 2r2^{-r}. We show that any learning algorithm for the learning problem corresponding to MM, with error, requires either a memory of size at least Ω(kε)\Omega\left(\frac{k \cdot \ell}{\varepsilon} \right), or at least 2Ω(r)2^{\Omega(r)} samples. In particular, this shows that for a large class of learning problems, same as those in [GRT'18], any learning algorithm requires either a memory of size at least Ω((logX)(logA)ε)\Omega\left(\frac{(\log |X|) \cdot (\log |A|)}{\varepsilon}\right) or an exponential number of noisy samples. Our proof is based on adapting the arguments in [Raz'17,GRT'18] to the noisy case.

Keywords

Cite

@article{arxiv.2107.02320,
  title  = {Memory-Sample Lower Bounds for Learning Parity with Noise},
  author = {Sumegha Garg and Pravesh K. Kothari and Pengda Liu and Ran Raz},
  journal= {arXiv preprint arXiv:2107.02320},
  year   = {2021}
}

Comments

19 pages. To appear in RANDOM 2021. arXiv admin note: substantial text overlap with arXiv:1708.02639

R2 v1 2026-06-24T03:54:55.911Z