Black-box Oracle Research Articles

Learning quantum symmetries with interactive quantum-classical variational algorithms

A symmetry of a state |ψ⟩ is a unitary operator of which |ψ⟩ is an eigenvector. When |ψ⟩ is an unknown state supplied by a black-box oracle, the state’s symmetries provide key physical insight into the quantum system; symmetries also boost many crucial quantum learning techniques. In this paper, we develop a variational hybrid quantum–classical learning scheme to systematically probe for symmetries of |ψ⟩ with no a priori assumptions about the state. This procedure can be used to learn various symmetries at the same time. In order to avoid re-learning already known symmetries, we introduce an interactive protocol with a classical deep neural net. The classical net thereby regularizes against repetitive findings and allows our algorithm to terminate empirically with all possible symmetries found. An iteration of the learning algorithm can be implemented efficiently with non-local SWAP gates; we also give a less efficient algorithm with only local operations, which may be more appropriate for current noisy quantum devices. We simulate our algorithm on representative families of states, including cluster states and ground states of Rydberg and Ising Hamiltonians. We also find that the numerical query complexity scales well for up to moderate system sizes.

Removal of SAT-Hard Instances in Logic Obfuscation Through Inference of Functionality

Logic obfuscation is a prominent approach to protect intellectual property within integrated circuits during fabrication. Many attacks on logic locking have been proposed, particularly in the Boolean satifiability (SAT) attack family, leading to the development of stronger obfuscation techniques. Some obfuscation techniques, including Full-Lock and InterLock, resist SAT attacks by inserting SAT-hard instances into the design, making the SAT attack infeasible. In this work, we observe that this class of obfuscation leaves most of the original design topology visible to an attacker, who can reverse-engineer the original design given the functionality of the SAT-hard instance. We show that an attacker can expose the SAT-hard instance functionality of Full-Lock or InterLock with a polynomial number of queries of its inputs and outputs. We then develop a mathematical framework showing how the functionality can be inferred using only a black-box oracle, as is commonly used in attacks in the literature. Using this framework, we develop a novel attack that allows a SAT-capable attacker to efficiently unlock designs obfuscated with Full-Lock. Our attack recovers the intellectual property from these obfuscation techniques that were previously thought secure. We empirically demonstrate the potency of our novel sensitization attack against benchmark circuits obfuscated with Full-Lock.

Inexact direct-search methods for bilevel optimization problems

In this work, we introduce new direct-search schemes for the solution of bilevel optimization (BO) problems. Our methods rely on a fixed accuracy blackbox oracle for the lower-level problem, and deal both with smooth and potentially nonsmooth true objectives. We thus analyze for the first time in the literature direct-search schemes in these settings, giving convergence guarantees to approximate stationary points, as well as complexity bounds in the smooth case. We also propose the first adaptation of mesh adaptive direct-search schemes for BO. Some preliminary numerical results on a standard set of bilevel optimization problems show the effectiveness of our new approaches.

GRASP: a goodness-of-fit test for classification learning

Abstract Performance of classifiers is often measured in terms of average accuracy on test data. Despite being a standard measure, average accuracy fails in characterising the fit of the model to the underlying conditional law of labels given the features vector (Y∣X), e.g. due to model misspecification, over fitting, and high-dimensionality. In this paper, we consider the fundamental problem of assessing the goodness-of-fit for a general binary classifier. Our framework does not make any parametric assumption on the conditional law Y∣X and treats that as a black-box oracle model which can be accessed only through queries. We formulate the goodness-of-fit assessment problem as a tolerance hypothesis testing of the form H0:E[Df(Bern(η(X))‖Bern(η^(X)))]≤τ where Df represents an f-divergence function, and η(x), η^(x), respectively, denote the true and an estimate likelihood for a feature vector x admitting a positive label. We propose a novel test, called Goodness-of-fit with Randomisation and Scoring Procedure (GRASP) for testing H0, which works in finite sample settings, no matter the features (distribution-free). We also propose model-X GRASP designed for model-X settings where the joint distribution of the features vector is known. Model-X GRASP uses this distributional information to achieve better power. We evaluate the performance of our tests through extensive numerical experiments.

Smoothed Online Optimization with Unreliable Predictions

We examine the problem of smoothed online optimization, where a decision maker must sequentially choose points in a normed vector space to minimize the sum of per-round, non-convex hitting costs and the costs of switching decisions between rounds. The decision maker has access to a black-box oracle, such as a machine learning model, that provides untrusted and potentially inaccurate predictions of the optimal decision in each round. The goal of the decision maker is to exploit the predictions if they are accurate, while guaranteeing performance that is not much worse than the hindsight optimal sequence of decisions, even when predictions are inaccurate. We impose the standard assumption that hitting costs are globally α-polyhedral. We propose a novel algorithm, Adaptive Online Switching (AOS), and prove that, for a large set of feasible δ > 0, it is (1+δ)-competitive if predictions are perfect, while also maintaining a uniformly bounded competitive ratio of 2~O (1/(α δ)) even when predictions are adversarial. Further, we prove that this trade-off is necessary and nearly optimal in the sense that any deterministic algorithm which is (1+δ)-competitive if predictions are perfect must be at least 2~Ω (1/(α δ)) -competitive when predictions are inaccurate. In fact, we observe a unique threshold-type behavior in this trade-off: if δ is not in the set of feasible options, then no algorithm is simultaneously (1 + δ)-competitive if predictions are perfect and ζ-competitive when predictions are inaccurate for any ζ < ∞. Furthermore, we discuss that memory is crucial in AOS by proving that any algorithm that does not use memory cannot benefit from predictions. We complement our theoretical results by a numerical study on a microgrid application.

Douglas–Rachford splitting and ADMM for nonconvex optimization: accelerated and Newton-type linesearch algorithms

Although the performance of popular optimization algorithms such as the Douglas–Rachford splitting (DRS) and the ADMM is satisfactory in convex and well-scaled problems, ill conditioning and nonconvexity pose a severe obstacle to their reliable employment. Expanding on recent convergence results for DRS and ADMM applied to nonconvex problems, we propose two linesearch algorithms to enhance and robustify these methods by means of quasi-Newton directions. The proposed algorithms are suited for nonconvex problems, require the same black-box oracle of DRS and ADMM, and maintain their (subsequential) convergence properties. Numerical evidence shows that the employment of L-BFGS in the proposed framework greatly improves convergence of DRS and ADMM, making them robust to ill conditioning. Under regularity and nondegeneracy assumptions at the limit point, superlinear convergence is shown when quasi-Newton Broyden directions are adopted.

Random Minibatch Subgradient Algorithms for Convex Problems with Functional Constraints

In this paper we consider non-smooth convex optimization problems with (possibly) infinite intersection of constraints. In contrast to the classical approach, where the constraints are usually represented as intersection of simple sets, which are easy to project onto, in this paper we consider that each constraint set is given as the level set of a convex but not necessarily differentiable function. For these settings we propose subgradient iterative algorithms with random minibatch feasibility updates. At each iteration, our algorithms take a subgradient step aimed at only minimizing the objective function and then a subsequent subgradient step minimizing the feasibility violation of the observed minibatch of constraints. The feasibility updates are performed based on either parallel or sequential random observations of several constraint components. We analyze the convergence behavior of the proposed algorithms for the case when the objective function is strongly convex and with bounded subgradients, while the functional constraints are endowed with a bounded first-order black-box oracle. For a diminishing stepsize, we prove sublinear convergence rates for the expected distances of the weighted averages of the iterates from the constraint set, as well as for the expected suboptimality of the function values along the weighted averages. Our convergence rates are known to be optimal for subgradient methods on this class of problems. Moreover, the rates depend explicitly on the minibatch size and show when minibatching helps a subgradient scheme with random feasibility updates.

Derivative-free optimization methods

In many optimization problems arising from scientific, engineering and artificial intelligence applications, objective and constraint functions are available only as the output of a black-box or simulation oracle that does not provide derivative information. Such settings necessitate the use of methods for derivative-free, or zeroth-order, optimization. We provide a review and perspectives on developments in these methods, with an emphasis on highlighting recent developments and on unifying treatment of such problems in the non-linear optimization and machine learning literature. We categorize methods based on assumed properties of the black-box functions, as well as features of the methods. We first overview the primary setting of deterministic methods applied to unconstrained, non-convex optimization problems where the objective function is defined by a deterministic black-box oracle. We then discuss developments in randomized methods, methods that assume some additional structure about the objective (including convexity, separability and general non-smooth compositions), methods for problems where the output of the black-box oracle is stochastic, and methods for handling different types of constraints.

Explaining AI Decisions Using Efficient Methods for Learning Sparse Boolean Formulae

In this paper, we consider the problem of learning Boolean formulae from examples obtained by actively querying an oracle that can label these examples as either positive or negative. This problem has received attention in both machine learning as well as formal methods communities, and it has been shown to have exponential worst-case complexity in the general case as well as for many restrictions. In this paper, we focus on learning sparse Boolean formulae which depend on only a small (but unknown) subset of the overall vocabulary of atomic propositions. We propose two algorithms—first, based on binary search in the Hamming space, and the second, based on random walk on the Boolean hypercube, to learn these sparse Boolean formulae with a given confidence. This assumption of sparsity is motivated by the problem of mining explanations for decisions made by artificially intelligent (AI) algorithms, where the explanation of individual decisions may depend on a small but unknown subset of all the inputs to the algorithm. We demonstrate the use of these algorithms in automatically generating explanations of these decisions. These explanations will make intelligent systems more understandable and accountable to human users, facilitate easier audits and provide diagnostic information in the case of failure. The proposed approach treats the AI algorithm as a black-box oracle; hence, it is broadly applicable and agnostic to the specific AI algorithm. We show that the number of examples needed for both proposed algorithms only grows logarithmically with the size of the vocabulary of atomic propositions. We illustrate the practical effectiveness of our approach on a diverse set of case studies.

Forward-Backward Envelope for the Sum of Two Nonconvex Functions: Further Properties and Nonmonotone Linesearch Algorithms

We propose ZeroFPR, a nonmonotone linesearch algorithm for minimizing the sum of two nonconvex functions, one of which is smooth and the other possibly nonsmooth. ZeroFPR is the first algorithm that, despite being fit for fully nonconvex problems and requiring only the black-box oracle of forward-backward splitting (FBS) --- namely evaluations of the gradient of the smooth term and of the proximity operator of the nonsmooth one --- achieves superlinear convergence rates under mild assumptions at the limit point when the linesearch directions satisfy a Dennis-Mor\'e condition, and we show that this is the case for quasi-Newton directions. Our approach is based on the forward-backward envelope (FBE), an exact and strictly continuous penalty function for the original cost. Extending previous results we show that, despite being nonsmooth for fully nonconvex problems, the FBE still enjoys favorable first- and second-order properties which are key for the convergence results of ZeroFPR. Our theoretical results are backed up by promising numerical simulations. On large-scale problems, by computing linesearch directions using limited-memory quasi-Newton updates our algorithm greatly outperforms FBS and its accelerated variant (AFBS).

Forward–backward quasi-Newton methods for nonsmooth optimization problems

The forward-backward splitting method (FBS) for minimizing a nonsmooth composite function can be interpreted as a (variable-metric) gradient method over a continuously differentiable function which we call forward-backward envelope (FBE). This allows to extend algorithms for smooth unconstrained optimization and apply them to nonsmooth (possibly constrained) problems. Since the FBE and its gradient can be computed by simply evaluating forward-backward steps, the resulting methods rely on the very same black-box oracle as FBS. We propose an algorithmic scheme that enjoys the same global convergence properties of FBS when the problem is convex, or when the objective function possesses the Kurdyka-{\L}ojasiewicz property at its critical points. Moreover, when using quasi-Newton directions the proposed method achieves superlinear convergence provided that usual second-order sufficiency conditions on the FBE hold at the limit point of the generated sequence. Such conditions translate into milder requirements on the original function involving generalized second-order differentiability. We show that BFGS fits our framework and that the limited-memory variant L-BFGS is well suited for large-scale problems, greatly outperforming FBS or its accelerated version in practice. The analysis of superlinear convergence is based on an extension of the Dennis and Mor\'e theorem for the proposed algorithmic scheme.

Obfuscating Re-encryption Algorithm With Flexible and Controllable Multi-Hop on Untrusted Outsourcing Server

An outsourcing re-encryption program can help a ciphertext owner (delegator) transform his/her ciphertext into another ciphertext of delegatee. For example, an e-mail receiver can re-transfer an encrypted e-mail to his secretary while allowing the e-mail to be readable for her. For a multi-hop re-encryption, the delegatee can re-encrypt the ciphertext to another user in delegation chain, repeatedly. Traditionally, this transformation is usually conducted by a proxy or an outsourcing server. However, the proxy or outsourcing server needs a re-encryption key (i.e., re-key) and the re-encryption program must execute in a black-box manner (cannot trace into or debug and monitor the program), and thus the outsource server must be semi-trusted. Actually, as the outsource program was run and fully controlled by the server, in this paper, we consider a stronger attack in the case that the re-encryption program was run on an untrusted/malicious server and even the server can trace into the codes and monitor the variables during the executing. We design a secure multi-hop re-encryption scheme, and then convert the re-encryption program into an obfuscated version with constant-hiding to ensure no sensitive information be revealed. The obfuscator of multi-hop re-encryption is to faithfully hide the program and its sensitive data that takes a re-encryption program/circuit as input and outputs another program with the same functionality, while revealing no more sensitive information (i.e., sensitive key and plaintext) than learns from the blackbox oracle access to the original program. We also present a flexible and controllable construction of re-encryption scheme, functionality model and its obfuscation version in leveled multilinear groups, and exemplify some scenarios to deploy in various applications. Finally, we provide the performance analysis of the obfuscator, such as functionality preservation of consistency, polynomial slowdown of performance, and average-case virtual black-box of security, and show that the obfuscator is efficient and practical in use.

An adaptive accelerated proximal gradient method and its homotopy continuation for sparse optimization

We consider optimization problems with an objective function that is the sum of two convex terms: one is smooth and given by a black-box oracle, and the other is general but with a simple, known structure. We first present an accelerated proximal gradient (APG) method for problems where the smooth part of the objective function is also strongly convex. This method incorporates an efficient line-search procedure, and achieves the optimal iteration complexity for such composite optimization problems. In case the strong convexity parameter is unknown, we also develop an adaptive scheme that can automatically estimate it on the fly, at the cost of a slightly worse iteration complexity. Then we focus on the special case of solving the $$\ell _1$$l1-regularized least-squares problem in the high-dimensional setting. In such a context, the smooth part of the objective (least-squares) is not strongly convex over the entire domain. Nevertheless, we can exploit its restricted strong convexity over sparse vectors using the adaptive APG method combined with a homotopy continuation scheme. We show that such a combination leads to a global geometric rate of convergence, and the overall iteration complexity has a weaker dependency on the restricted condition number than previous work.

On lower complexity bounds for large-scale smooth convex optimization

We derive lower bounds on the black-box oracle complexity of large-scale smooth convex minimization problems, with emphasis on minimizing smooth (with Hölder continuous, with a given exponent and constant, gradient) convex functions over high-dimensional ‖⋅‖p-balls, 1≤p≤∞. Our bounds turn out to be tight (up to logarithmic in the design dimension factors), and can be viewed as a substantial extension of the existing lower complexity bounds for large-scale convex minimization covering the nonsmooth case and the “Euclidean” smooth case (minimization of convex functions with Lipschitz continuous gradients over Euclidean balls). As a byproduct of our results, we demonstrate that the classical Conditional Gradient algorithm is near-optimal, in the sense of Information-Based Complexity Theory, when minimizing smooth convex functions over high-dimensional ‖⋅‖∞-balls and their matrix analogies–spectral norm balls in the spaces of square matrices.

Distribution-Aware Sampling and Weighted Model Counting for SAT

Given a CNF formula and a weight for each assignment of values tovariables, two natural problems are weighted model counting anddistribution-aware sampling of satisfying assignments. Both problems have a wide variety of important applications. Due to the inherentcomplexity of the exact versions of the problems, interest has focusedon solving them approximately. Prior work in this area scaled only tosmall problems in practice, or failed to provide strong theoreticalguarantees, or employed a computationally-expensive most-probable-explanation ({\MPE}) queries that assumes prior knowledge of afactored representation of the weight distribution. We identify a novel parameter,\emph{tilt}, which is the ratio of the maximum weight of satisfying assignment to minimum weightof satisfying assignment and present anovel approach that works with a black-box oracle for weights ofassignments and requires only an {\NP}-oracle (in practice, a {\SAT}-solver) to solve both thecounting and sampling problems when the tilt is small. Our approach provides strong theoretical guarantees, and scales toproblems involving several thousand variables. We also show that theassumption of small tilt can be significantly relaxed while improving computational efficiency if a factored representation of the weights is known.

Quantum inference on Bayesian networks

Performing exact inference on Bayesian networks is known to be $#P$-hard. Typically approximate inference techniques are used instead to sample from the distribution on query variables given the values $e$ of evidence variables. Classically, a single unbiased sample is obtained from a Bayesian network on $n$ variables with at most $m$ parents per node in time $O(nmP{(e)}^{\ensuremath{-}1})$, depending critically on $P(e)$, the probability that the evidence might occur in the first place. By implementing a quantum version of rejection sampling, we obtain a square-root speedup, taking $O(n{2}^{m}P{(e)}^{\ensuremath{-}\frac{1}{2}})$ time per sample. We exploit the Bayesian network's graph structure to efficiently construct a quantum state, a q-sample, representing the intended classical distribution, and also to efficiently apply amplitude amplification, the source of our speedup. Thus, our speedup is notable as it is unrelativized---we count primitive operations and require no blackbox oracle queries.

Gradient methods for minimizing composite functions

In this paper we analyze several new methods for solving optimization problems with the objective function formed as a sum of two terms: one is smooth and given by a black-box oracle, and another is a simple general convex function with known structure. Despite the absence of good properties of the sum, such problems, both in convex and nonconvex cases, can be solved with efficiency typical for the first part of the objective. For convex problems of the above structure, we consider primal and dual variants of the gradient method (with convergence rate $$O\left({1 \over k}\right)$$ ), and an accelerated multistep version with convergence rate $$O\left({1 \over k^2}\right)$$ , where $$k$$ is the iteration counter. For nonconvex problems with this structure, we prove convergence to a point from which there is no descent direction. In contrast, we show that for general nonsmooth, nonconvex problems, even resolving the question of whether a descent direction exists from a point is NP-hard. For all methods, we suggest some efficient “line search” procedures and show that the additional computational work necessary for estimating the unknown problem class parameters can only multiply the complexity of each iteration by a small constant factor. We present also the results of preliminary computational experiments, which confirm the superiority of the accelerated scheme.

Improved algorithms for quantum identification of Boolean oracles

The oracle identification problem (OIP) was introduced by Ambainis et al. [A. Ambainis, K. Iwama, A. Kawachi, H. Masuda, R.H. Putra, S. Yamashita, Quantum identification of boolean oracles, in: Proc. of STACS’04, in: LNCS, vol. 2996, 2004, pp. 105–116]. It is given as a set S of M oracles and a blackbox oracle f . Our task is to figure out which oracle in S is equal to the blackbox f by making queries to f . OIP includes several problems such as the Grover Search as special cases. In this paper, we improve the algorithms in [A. Ambainis, K. Iwama, A. Kawachi, H. Masuda, R.H. Putra, S. Yamashita, Quantum identification of boolean oracles, in: Proc. of STACS’04, in: LNCS, vol. 2996, 2004, pp. 105–116] by providing a mostly optimal upper bound of query complexity for this problem: (i) For any oracle set S such that | S | ≤ 2 N d ( d < 1 ) , we design an algorithm whose query complexity is O ( N log M / log N ) , matching the lower bound proved in [A. Ambainis, K. Iwama, A. Kawachi, H. Masuda, R.H. Putra, S. Yamashita, Quantum identification of boolean oracles, in: Proc. of STACS’04, in: LNCS, vol. 2996, 2004, pp. 105–116]. (ii) Our algorithm also works for the range between 2 N d and 2 N / log N (where the bound becomes O ( N ) ), but the gap between the upper and lower bounds worsens gradually. (iii) Our algorithm is robust, namely, it exhibits the same performance (up to a constant factor) against noisy oracles as also shown in the literature [M. Adcock, R. Cleve, A quantum Goldreich–Levin theorem with cryptographic applications, in: Proc. of STACS’02, in: LNCS, vol. 2285, 2002, pp. 323–334; H. Buhrman, I. Newman, H. Röhrig, R. deWolf, Robust quantum algorithms and polynomials, in: Proc. of STACS’05, in: LNCS, vol. 3404, 2005, pp. 593–604; P. Høyer, M. Mosca, R. de Wolf, Quantum search on bounded-error inputs, in: Proc. of ICALP’03, in: LNCS, vol. 2719, 2003, pp. 291–299] for special cases of OIP.

The Quantum Complexity of Set Membership

Abstract. We study the quantum complexity of the static set membership problem: given a subset S (|S| ≤ n ) of a universe of size m ( >> n ), store it as a table, T: {0,1} r --> {0,1} , of bits so that queries of the form ``Is x in S ?'' can be answered. The goal is to use a small table and yet answer queries using few bit probes. This problem was considered recently by Buhrman et al. [BMRV], who showed lower and upper bounds for this problem in the classical deterministic and randomized models. In this paper we formulate this problem in the ``quantum bit probe model''. We assume that access to the table T is provided by means of a black box (oracle) unitary transform O T that takes the basis state | y,b > to the basis state | y,b \oplus T(y) > . The query algorithm is allowed to apply O T on any superposition of basis states. We show tradeoff results between space (defined as 2 r ) and number of probes (oracle calls) in this model. Our results show that the lower bounds shown in [BMRV] for the classical model also hold (with minor differences) in the quantum bit probe model. These bounds almost match the classical upper bounds. Our lower bounds are proved using linear algebraic arguments.