Properly Learning Poisson Binomial Distributions in Almost Polynomial Time

I. Diakonikolas, D. M. Kane, A. Stewart
; 29th Annual Conference on Learning Theory, PMLR 49:850-878, 2016.

Abstract

We give an algorithm for properly learning Poisson binomial distributions. A Poisson binomial distribution (PBD) of order n ∈\mathbbZ_+ is the discrete probability distribution of the sum of n mutually independent Bernoulli random variables. Given \widetildeO(1/ε^2) samples from an unknown PBD P, our algorithm runs in time (1/\eps)^O(\log \log (1/ε)), and outputs a hypothesis PBD that is ε-close to P in total variation distance. The sample complexity of our algorithm is known to be nearly-optimal, up to logarithmic factors, as established in previous work. However, the previously best known running time for properly learning PBDs was (1/ε)^O(\log(1/ε)), and was essentially obtained by enumeration over an appropriate ε-cover. We remark that the running time of this cover-based approach cannot be improved, as any ε-cover for the space of PBDs has size (1/ε)^Ω(\log(1/ε)). As one of our main contributions, we provide a novel structural characterization of PBDs, showing that any PBD P is ε-close to another PBD Q with O(\log(1/ε)) distinct parameters. More precisely, we prove that, for all ε>0, there exists an explicit collection \calM of (1/ε)^O(\log \log (1/ε)) vectors of multiplicities, such that for any PBD P there exists a PBD Q with O(\log(1/ε)) distinct parameters whose multiplicities are given by some element of \cal M, such that Q is ε-close to P. Our proof combines tools from Fourier analysis and algebraic geometry. Our approach to the proper learning problem is as follows: Starting with an accurate non-proper hypothesis, we fit a PBD to this hypothesis. This fitting problem can be formulated as a natural polynomial optimization problem. Our aforementioned structural characterization allows us to reduce the corresponding fitting problem to a collection of (1/ε)^O(\log \log(1/ε)) systems of low-degree polynomial inequalities. We show that each such system can be solved in time (1/ε)^O(\log \log(1/ε)), which yields the overall running time of our algorithm.

Cite this Paper


BibTeX
@InProceedings{pmlr-v49-diakonikolas16b, title = {Properly Learning Poisson Binomial Distributions in Almost Polynomial Time}, author = {I. Diakonikolas and D. M. Kane and A. Stewart}, booktitle = {29th Annual Conference on Learning Theory}, pages = {850--878}, year = {2016}, editor = {Vitaly Feldman and Alexander Rakhlin and Ohad Shamir}, volume = {49}, series = {Proceedings of Machine Learning Research}, address = {Columbia University, New York, New York, USA}, month = {23--26 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v49/diakonikolas16b.pdf}, url = {http://proceedings.mlr.press/v49/diakonikolas16b.html}, abstract = {We give an algorithm for properly learning Poisson binomial distributions. A Poisson binomial distribution (PBD) of order n ∈\mathbbZ_+ is the discrete probability distribution of the sum of n mutually independent Bernoulli random variables. Given \widetildeO(1/ε^2) samples from an unknown PBD P, our algorithm runs in time (1/\eps)^O(\log \log (1/ε)), and outputs a hypothesis PBD that is ε-close to P in total variation distance. The sample complexity of our algorithm is known to be nearly-optimal, up to logarithmic factors, as established in previous work. However, the previously best known running time for properly learning PBDs was (1/ε)^O(\log(1/ε)), and was essentially obtained by enumeration over an appropriate ε-cover. We remark that the running time of this cover-based approach cannot be improved, as any ε-cover for the space of PBDs has size (1/ε)^Ω(\log(1/ε)). As one of our main contributions, we provide a novel structural characterization of PBDs, showing that any PBD P is ε-close to another PBD Q with O(\log(1/ε)) distinct parameters. More precisely, we prove that, for all ε>0, there exists an explicit collection \calM of (1/ε)^O(\log \log (1/ε)) vectors of multiplicities, such that for any PBD P there exists a PBD Q with O(\log(1/ε)) distinct parameters whose multiplicities are given by some element of \cal M, such that Q is ε-close to P. Our proof combines tools from Fourier analysis and algebraic geometry. Our approach to the proper learning problem is as follows: Starting with an accurate non-proper hypothesis, we fit a PBD to this hypothesis. This fitting problem can be formulated as a natural polynomial optimization problem. Our aforementioned structural characterization allows us to reduce the corresponding fitting problem to a collection of (1/ε)^O(\log \log(1/ε)) systems of low-degree polynomial inequalities. We show that each such system can be solved in time (1/ε)^O(\log \log(1/ε)), which yields the overall running time of our algorithm. } }
Endnote
%0 Conference Paper %T Properly Learning Poisson Binomial Distributions in Almost Polynomial Time %A I. Diakonikolas %A D. M. Kane %A A. Stewart %B 29th Annual Conference on Learning Theory %C Proceedings of Machine Learning Research %D 2016 %E Vitaly Feldman %E Alexander Rakhlin %E Ohad Shamir %F pmlr-v49-diakonikolas16b %I PMLR %J Proceedings of Machine Learning Research %P 850--878 %U http://proceedings.mlr.press %V 49 %W PMLR %X We give an algorithm for properly learning Poisson binomial distributions. A Poisson binomial distribution (PBD) of order n ∈\mathbbZ_+ is the discrete probability distribution of the sum of n mutually independent Bernoulli random variables. Given \widetildeO(1/ε^2) samples from an unknown PBD P, our algorithm runs in time (1/\eps)^O(\log \log (1/ε)), and outputs a hypothesis PBD that is ε-close to P in total variation distance. The sample complexity of our algorithm is known to be nearly-optimal, up to logarithmic factors, as established in previous work. However, the previously best known running time for properly learning PBDs was (1/ε)^O(\log(1/ε)), and was essentially obtained by enumeration over an appropriate ε-cover. We remark that the running time of this cover-based approach cannot be improved, as any ε-cover for the space of PBDs has size (1/ε)^Ω(\log(1/ε)). As one of our main contributions, we provide a novel structural characterization of PBDs, showing that any PBD P is ε-close to another PBD Q with O(\log(1/ε)) distinct parameters. More precisely, we prove that, for all ε>0, there exists an explicit collection \calM of (1/ε)^O(\log \log (1/ε)) vectors of multiplicities, such that for any PBD P there exists a PBD Q with O(\log(1/ε)) distinct parameters whose multiplicities are given by some element of \cal M, such that Q is ε-close to P. Our proof combines tools from Fourier analysis and algebraic geometry. Our approach to the proper learning problem is as follows: Starting with an accurate non-proper hypothesis, we fit a PBD to this hypothesis. This fitting problem can be formulated as a natural polynomial optimization problem. Our aforementioned structural characterization allows us to reduce the corresponding fitting problem to a collection of (1/ε)^O(\log \log(1/ε)) systems of low-degree polynomial inequalities. We show that each such system can be solved in time (1/ε)^O(\log \log(1/ε)), which yields the overall running time of our algorithm.
RIS
TY - CPAPER TI - Properly Learning Poisson Binomial Distributions in Almost Polynomial Time AU - I. Diakonikolas AU - D. M. Kane AU - A. Stewart BT - 29th Annual Conference on Learning Theory PY - 2016/06/06 DA - 2016/06/06 ED - Vitaly Feldman ED - Alexander Rakhlin ED - Ohad Shamir ID - pmlr-v49-diakonikolas16b PB - PMLR SP - 850 DP - PMLR EP - 878 L1 - http://proceedings.mlr.press/v49/diakonikolas16b.pdf UR - http://proceedings.mlr.press/v49/diakonikolas16b.html AB - We give an algorithm for properly learning Poisson binomial distributions. A Poisson binomial distribution (PBD) of order n ∈\mathbbZ_+ is the discrete probability distribution of the sum of n mutually independent Bernoulli random variables. Given \widetildeO(1/ε^2) samples from an unknown PBD P, our algorithm runs in time (1/\eps)^O(\log \log (1/ε)), and outputs a hypothesis PBD that is ε-close to P in total variation distance. The sample complexity of our algorithm is known to be nearly-optimal, up to logarithmic factors, as established in previous work. However, the previously best known running time for properly learning PBDs was (1/ε)^O(\log(1/ε)), and was essentially obtained by enumeration over an appropriate ε-cover. We remark that the running time of this cover-based approach cannot be improved, as any ε-cover for the space of PBDs has size (1/ε)^Ω(\log(1/ε)). As one of our main contributions, we provide a novel structural characterization of PBDs, showing that any PBD P is ε-close to another PBD Q with O(\log(1/ε)) distinct parameters. More precisely, we prove that, for all ε>0, there exists an explicit collection \calM of (1/ε)^O(\log \log (1/ε)) vectors of multiplicities, such that for any PBD P there exists a PBD Q with O(\log(1/ε)) distinct parameters whose multiplicities are given by some element of \cal M, such that Q is ε-close to P. Our proof combines tools from Fourier analysis and algebraic geometry. Our approach to the proper learning problem is as follows: Starting with an accurate non-proper hypothesis, we fit a PBD to this hypothesis. This fitting problem can be formulated as a natural polynomial optimization problem. Our aforementioned structural characterization allows us to reduce the corresponding fitting problem to a collection of (1/ε)^O(\log \log(1/ε)) systems of low-degree polynomial inequalities. We show that each such system can be solved in time (1/ε)^O(\log \log(1/ε)), which yields the overall running time of our algorithm. ER -
APA
Diakonikolas, I., Kane, D.M. & Stewart, A.. (2016). Properly Learning Poisson Binomial Distributions in Almost Polynomial Time. 29th Annual Conference on Learning Theory, in PMLR 49:850-878

Related Material