Provable Advantage in Quantum PAC Learning

• URL: http://arxiv.org/abs/2309.10887v1
• Date: Tue, 19 Sep 2023 19:26:20 GMT
• Title: Provable Advantage in Quantum PAC Learning
• Authors: Wilfred Salmon, Sergii Strelchuk, Tom Gur
• Abstract summary: We show that PAC learners can achieve a generic advantage in quantum learning. Up to polylogarithmic factors, this is a square root improvement over the classical learning sample complexity.
• Score: 3.291862617649511
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We revisit the problem of characterising the complexity of Quantum PAC learning, as introduced by Bshouty and Jackson [SIAM J. Comput. 1998, 28, 1136-1153]. Several quantum advantages have been demonstrated in this setting, however, none are generic: they apply to particular concept classes and typically only work when the distribution that generates the data is known. In the general case, it was recently shown by Arunachalam and de Wolf [JMLR, 19 (2018) 1-36] that quantum PAC learners can only achieve constant factor advantages over classical PAC learners. We show that with a natural extension of the definition of quantum PAC learning used by Arunachalam and de Wolf, we can achieve a generic advantage in quantum learning. To be precise, for any concept class $\mathcal{C}$ of VC dimension $d$, we show there is an $(\epsilon, \delta)$-quantum PAC learner with sample complexity \[ O\left(\frac{1}{\sqrt{\epsilon}}\left[d+ \log(\frac{1}{\delta})\right]\log^9(1/\epsilon)\right). \] Up to polylogarithmic factors, this is a square root improvement over the classical learning sample complexity. We show the tightness of our result by proving an $\Omega(d/\sqrt{\epsilon})$ lower bound that matches our upper bound up to polylogarithmic factors.

Related papers

• New Bounds on Quantum Sample Complexity of Measurement Classes [18.531114735719274]
  This paper studies quantum supervised learning for classical inference from quantum states. The hardness of learning is measured via sample complexity under a quantum counterpart of the well-known probably approximately correct (PAC)
  arXiv  Detail & Related papers  (2024-08-22T18:43:13Z)
• Fast Rates for Bandit PAC Multiclass Classification [73.17969992976501]
  We study multiclass PAC learning with bandit feedback, where inputs are classified into one of $K$ possible labels and feedback is limited to whether or not the predicted labels are correct. Our main contribution is in designing a novel learning algorithm for the agnostic $(varepsilon,delta)$PAC version of the problem.
  arXiv  Detail & Related papers  (2024-06-18T08:54:04Z)
• Proper vs Improper Quantum PAC learning [3.292420364429101]
  We present an algorithm for the Quantum Coupon Collector problem with sample complexity. We prove that its sample complexity matches that of the classical Coupon Collector problem for both modes of learning. We hope that padding can more generally lift other forms of classical learning behaviour to quantum setting.
  arXiv  Detail & Related papers  (2024-03-05T19:49:44Z)
• The Power of Unentangled Quantum Proofs with Non-negative Amplitudes [55.90795112399611]
  We study the power of unentangled quantum proofs with non-negative amplitudes, a class which we denote $textQMA+(2)$. In particular, we design global protocols for small set expansion, unique games, and PCP verification. We show that QMA(2) is equal to $textQMA+(2)$ provided the gap of the latter is a sufficiently large constant.
  arXiv  Detail & Related papers  (2024-02-29T01:35:46Z)
• A Quadratic Sample Complexity Reduction for Agnostic Learning via Quantum Algorithms [0.0]
  We find a new sample complexity upper bound of $O(mboxlog(frac1delta)/epsilon) as $epsilon,deltarightarrow 0arrow. For general learning, the quantum speedup in the rate of learning is quadratic in $epsilon-1$.
  arXiv  Detail & Related papers  (2023-10-24T07:39:16Z)
• Optimal lower bounds for Quantum Learning via Information Theory [3.093890460224435]
  We derive optimal lower bounds for quantum sample complexity in both the PAC and models via an information-theoretic approach. We then turn to a quantum analogue of the Coupon Collector problem, a classic problem from probability theory. All the aspects of the Quantum Coupon Collector problem we study rest on properties of the spectrum of the associated Gram matrix.
  arXiv  Detail & Related papers  (2023-01-05T18:55:04Z)
• Quantum Depth in the Random Oracle Model [57.663890114335736]
  We give a comprehensive characterization of the computational power of shallow quantum circuits combined with classical computation. For some problems, the ability to perform adaptive measurements in a single shallow quantum circuit is more useful than the ability to perform many shallow quantum circuits without adaptive measurements.
  arXiv  Detail & Related papers  (2022-10-12T17:54:02Z)
• Entanglement and coherence in Bernstein-Vazirani algorithm [58.720142291102135]
  Bernstein-Vazirani algorithm allows one to determine a bit string encoded into an oracle. We analyze in detail the quantum resources in the Bernstein-Vazirani algorithm. We show that in the absence of entanglement, the performance of the algorithm is directly related to the amount of quantum coherence in the initial state.
  arXiv  Detail & Related papers  (2022-05-26T20:32:36Z)
• Exponential separations between learning with and without quantum memory [17.763817187554096]
  We study the power of quantum memory for learning properties of quantum systems and dynamics. Many state-of-the-art learning algorithms require access to an additional external quantum memory. We show that this trade-off is inherent in a wide range of learning problems.
  arXiv  Detail & Related papers  (2021-11-10T19:03:49Z)
• Algorithms and SQ Lower Bounds for PAC Learning One-Hidden-Layer ReLU Networks [48.32532049640782]
  We study the problem of learning one-hidden-layer ReLU networks with $k$ hidden units on $mathbbRd$ under Gaussian marginals. For the case of positive coefficients, we give the first-time algorithm for this learning problem for $k$ up to $tildeOOmega(sqrtlog d)$.
  arXiv  Detail & Related papers  (2020-06-22T17:53:54Z)
• Quantum Coupon Collector [62.58209964224025]
  We study how efficiently a $k$-element set $Ssubseteq[n]$ can be learned from a uniform superposition $|Srangle of its elements. We give tight bounds on the number of quantum samples needed for every $k$ and $n$, and we give efficient quantum learning algorithms.
  arXiv  Detail & Related papers  (2020-02-18T16:14:55Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.