arXiv subjects

Li-Yang Tan

Publications and source records attributed to Li-Yang Tan.

At least 37 records · Page 2Linked to original sources

arXiv2022

On the power of adaptivity in statistical adversaries

We study a fundamental question concerning adversarial noise models in statistical problems where the algorithm receives i.i.d. draws from a distribution $\mathcal{D}$. The definitions of these adversaries specify the type of allowable corruptions (noise model) as well as when these corruptions can be made (adaptivity); the latter differentiates between oblivious adversaries that can only corrupt the distribution $\mathcal{D}$ and adaptive adversaries that can have their corruptions depend on the specific sample $S$ that is drawn from $\mathcal{D}$. In this work, we investigate whether oblivious adversaries are effectively equivalent to adaptive adversaries, across all noise models studied in the literature. Specifically, can the behavior of an algorithm $\mathcal{A}$ in the presence of oblivious adversaries always be well-approximated by that of an algorithm $\mathcal{A}'$ in the presence of adaptive adversaries? Our first result shows that this is indeed the case for the broad class of statistical query algorithms, under all reasonable noise models. We then show that in the specific case of additive noise, this equivalence holds for all algorithms. Finally, we map out an approach towards proving this statement in its fullest generality, for all algorithms and under all reasonable noise models.

cs.LG↗

arXiv2022

Open Problem: Properly learning decision trees in polynomial time?

The authors recently gave an $n^{O(\log\log n)}$ time membership query algorithm for properly learning decision trees under the uniform distribution (Blanc et al., 2021). The previous fastest algorithm for this problem ran in $n^{O(\log n)}$ time, a consequence of Ehrenfeucht and Haussler (1989)'s classic algorithm for the distribution-free setting. In this article we highlight the natural open problem of obtaining a polynomial-time algorithm, discuss possible avenues towards obtaining it, and state intermediate milestones that we believe are of independent interest.

cs.DS↗

arXiv2022

Popular decision tree algorithms are provably noise tolerant

Using the framework of boosting, we prove that all impurity-based decision tree learning algorithms, including the classic ID3, C4.5, and CART, are highly noise tolerant. Our guarantees hold under the strongest noise model of nasty noise, and we provide near-matching upper and lower bounds on the allowable noise rate. We further show that these algorithms, which are simple and have long been central to everyday machine learning, enjoy provable guarantees in the noisy setting that are unmatched by existing algorithms in the theoretical literature on decision tree learning. Taken together, our results add to an ongoing line of research that seeks to place the empirical success of these practical decision tree algorithms on firm theoretical footing.

cs.LG↗

arXiv2022

Reconstructing decision trees

We give the first {\sl reconstruction algorithm} for decision trees: given queries to a function $f$ that is $\mathrm{opt}$-close to a size-$s$ decision tree, our algorithm provides query access to a decision tree $T$ where: $\circ$ $T$ has size $S = s^{O((\log s)^2/\varepsilon^3)}$; $\circ$ $\mathrm{dist}(f,T)\le O(\mathrm{opt})+\varepsilon$; $\circ$ Every query to $T$ is answered with $\mathrm{poly}((\log s)/\varepsilon)\cdot \log n$ queries to $f$ and in $\mathrm{poly}((\log s)/\varepsilon)\cdot n\log n$ time. This yields a {\sl tolerant tester} that distinguishes functions that are close to size-$s$ decision trees from those that are far from size-$S$ decision trees. The polylogarithmic dependence on $s$ in the efficiency of our tester is exponentially smaller than that of existing testers. Since decision tree complexity is well known to be related to numerous other boolean function properties, our results also provide a new algorithms for reconstructing and testing these properties.

cs.DS↗

arXiv2022

The composition complexity of majority

We study the complexity of computing majority as a composition of local functions: \[ \text{Maj}_n = h(g_1,\ldots,g_m), \] where each $g_j :\{0,1\}^{n} \to \{0,1\}$ is an arbitrary function that queries only $k \ll n$ variables and $h : \{0,1\}^m \to \{0,1\}$ is an arbitrary combining function. We prove an optimal lower bound of \[ m \ge Ω\left( \frac{n}{k} \log k \right) \] on the number of functions needed, which is a factor $Ω(\log k)$ larger than the ideal $m = n/k$. We call this factor the composition overhead; previously, no superconstant lower bounds on it were known for majority. Our lower bound recovers, as a corollary and via an entirely different proof, the best known lower bound for bounded-width branching programs for majority (Alon and Maass '86, Babai et al. '90). It is also the first step in a plan that we propose for breaking a longstanding barrier in lower bounds for small-depth boolean circuits. Novel aspects of our proof include sharp bounds on the information lost as computation flows through the inner functions $g_j$, and the bootstrapping of lower bounds for a multi-output function (Hamming weight) into lower bounds for a single-output one (majority).

cs.CC↗

arXiv2022

Almost 3-Approximate Correlation Clustering in Constant Rounds

We study parallel algorithms for correlation clustering. Each pair among $n$ objects is labeled as either "similar" or "dissimilar". The goal is to partition the objects into arbitrarily many clusters while minimizing the number of disagreements with the labels. Our main result is an algorithm that for any $ε> 0$ obtains a $(3+ε)$-approximation in $O(1/ε)$ rounds (of models such as massively parallel computation, local, and semi-streaming). This is a culminating point for the rich literature on parallel correlation clustering. On the one hand, the approximation (almost) matches a natural barrier of 3 for combinatorial algorithms. On the other hand, the algorithm's round-complexity is essentially constant. To achieve this result, we introduce a simple $O(1/ε)$-round parallel algorithm. Our main result is to provide an analysis of this algorithm, showing that it achieves a $(3+ε)$-approximation. Our analysis draws on new connections to sublinear-time algorithms. Specifically, it builds on the work of Yoshida, Yamamoto, and Ito [STOC'09] on bounding the "query complexity" of greedy maximal independent set. To our knowledge, this is the first application of this method in analyzing the approximation ratio of any algorithm.

cs.DS↗

arXiv2022

Tradeoffs for small-depth Frege proofs

We study the complexity of small-depth Frege proofs and give the first tradeoffs between the size of each line and the number of lines. Existing lower bounds apply to the overall proof size -- the sum of sizes of all lines -- and do not distinguish between these notions of complexity. For depth-$d$ Frege proofs of the Tseitin principle on the $n \times n$ grid where each line is a size-$s$ formula, we prove that $\exp(n/2^{Ω(d\sqrt{\log s})})$ many lines are necessary. This yields new lower bounds on line complexity that are not implied by Håstad's recent $\exp(n^{Ω(1/d)})$ lower bound on the overall proof size. For $s = \mathrm{poly}(n)$, for example, our lower bound remains $\exp(n^{1-o(1)})$ for all $d = o(\sqrt{\log n})$, whereas Håstad's lower bound is $\exp(n^{o(1)})$ once $d = ω_n(1)$. Our main conceptual contribution is the simple observation that techniques for establishing correlation bounds in circuit complexity can be leveraged to establish such tradeoffs in proof complexity.

cs.CC↗

arXiv2022

The Query Complexity of Certification

We study the problem of {\sl certification}: given queries to a function $f : \{0,1\}^n \to \{0,1\}$ with certificate complexity $\le k$ and an input $x^\star$, output a size-$k$ certificate for $f$'s value on $x^\star$. This abstractly models a central problem in explainable machine learning, where we think of $f$ as a blackbox model that we seek to explain the predictions of. For monotone functions, a classic local search algorithm of Angluin accomplishes this task with $n$ queries, which we show is optimal for local search algorithms. Our main result is a new algorithm for certifying monotone functions with $O(k^8 \log n)$ queries, which comes close to matching the information-theoretic lower bound of $Ω(k \log n)$. The design and analysis of our algorithm are based on a new connection to threshold phenomena in monotone functions. We further prove exponential-in-$k$ lower bounds when $f$ is non-monotone, and when $f$ is monotone but the algorithm is only given random examples of $f$. These lower bounds show that assumptions on the structure of $f$ and query access to it are both necessary for the polynomial dependence on $k$ that we achieve.

cs.DS↗

arXiv2022

Fooling Gaussian PTFs via Local Hyperconcentration

We give a pseudorandom generator that fools degree-$d$ polynomial threshold functions over $n$-dimensional Gaussian space with seed length $\mathrm{poly}(d)\cdot \log n$. All previous generators had a seed length with at least a $2^d$ dependence on $d$. The key new ingredient is a Local Hyperconcentration Theorem, which shows that every degree-$d$ Gaussian polynomial is hyperconcentrated almost everywhere at scale $d^{-O(1)}$.

cs.CC↗

arXiv2021

Properly learning decision trees in almost polynomial time

We give an $n^{O(\log\log n)}$-time membership query algorithm for properly and agnostically learning decision trees under the uniform distribution over $\{\pm 1\}^n$. Even in the realizable setting, the previous fastest runtime was $n^{O(\log n)}$, a consequence of a classic algorithm of Ehrenfeucht and Haussler. Our algorithm shares similarities with practical heuristics for learning decision trees, which we augment with additional ideas to circumvent known lower bounds against these heuristics. To analyze our algorithm, we prove a new structural result for decision trees that strengthens a theorem of O'Donnell, Saks, Schramm, and Servedio. While the OSSS theorem says that every decision tree has an influential variable, we show how every decision tree can be "pruned" so that every variable in the resulting tree is influential.

cs.DS↗

arXiv2021

Provably efficient, succinct, and precise explanations

We consider the problem of explaining the predictions of an arbitrary blackbox model $f$: given query access to $f$ and an instance $x$, output a small set of $x$'s features that in conjunction essentially determines $f(x)$. We design an efficient algorithm with provable guarantees on the succinctness and precision of the explanations that it returns. Prior algorithms were either efficient but lacked such guarantees, or achieved such guarantees but were inefficient. We obtain our algorithm via a connection to the problem of {\sl implicitly} learning decision trees. The implicit nature of this learning task allows for efficient algorithms even when the complexity of $f$ necessitates an intractably large surrogate decision tree. We solve the implicit learning problem by bringing together techniques from learning theory, local computation algorithms, and complexity theory. Our approach of "explaining by implicit learning" shares elements of two previously disparate methods for post-hoc explanations, global and local explanations, and we make the case that it enjoys advantages of both.

cs.LG↗

arXiv2021

Sharper bounds on the Fourier concentration of DNFs

In 1992 Mansour proved that every size-$s$ DNF formula is Fourier-concentrated on $s^{O(\log\log s)}$ coefficients. We improve this to $s^{O(\log\log k)}$ where $k$ is the read number of the DNF. Since $k$ is always at most $s$, our bound matches Mansour's for all DNFs and strengthens it for small-read ones. The previous best bound for read-$k$ DNFs was $s^{O(k^{3/2})}$. For $k$ up to $\tildeΘ(\log\log s)$, we further improve our bound to the optimal $\mathrm{poly}(s)$; previously no such bound was known for any $k = ω_s(1)$. Our techniques involve new connections between the term structure of a DNF, viewed as a set system, and its Fourier spectrum.

cs.CC↗

arXiv2021

Decision tree heuristics can fail, even in the smoothed setting

Greedy decision tree learning heuristics are mainstays of machine learning practice, but theoretical justification for their empirical success remains elusive. In fact, it has long been known that there are simple target functions for which they fail badly (Kearns and Mansour, STOC 1996). Recent work of Brutzkus, Daniely, and Malach (COLT 2020) considered the smoothed analysis model as a possible avenue towards resolving this disconnect. Within the smoothed setting and for targets $f$ that are $k$-juntas, they showed that these heuristics successfully learn $f$ with depth-$k$ decision tree hypotheses. They conjectured that the same guarantee holds more generally for targets that are depth-$k$ decision trees. We provide a counterexample to this conjecture: we construct targets that are depth-$k$ decision trees and show that even in the smoothed setting, these heuristics build trees of depth $2^{Ω(k)}$ before achieving high accuracy. We also show that the guarantees of Brutzkus et al. cannot extend to the agnostic setting: there are targets that are very close to $k$-juntas, for which these heuristics build trees of depth $2^{Ω(k)}$ before achieving high accuracy.

cs.LG↗

arXiv2021

On the Power and Limitations of Branch and Cut

The Stabbing Planes proof system was introduced to model the reasoning carried out in practical mixed integer programming solvers. As a proof system, it is powerful enough to simulate Cutting Planes and to refute the Tseitin formulas -- certain unsatisfiable systems of linear equations mod 2 -- which are canonical hard examples for many algebraic proof systems. In a recent (and surprising) result, Dadush and Tiwari showed that these short refutations of the Tseitin formulas could be translated into quasi-polynomial size and depth Cutting Planes proofs, refuting a long-standing conjecture. This translation raises several interesting questions. First, whether all Stabbing Planes proofs can be efficiently simulated by Cutting Planes. This would allow for the substantial analysis done on the Cutting Planes system to be lifted to practical mixed integer programming solvers. Second, whether the quasi-polynomial depth of these proofs is inherent to Cutting Planes. In this paper we make progress towards answering both of these questions. First, we show that any Stabbing Planes proof with bounded coefficients SP* can be translated into Cutting Planes. As a consequence of the known lower bounds for Cutting Planes, this establishes the first exponential lower bounds on SP*. Using this translation, we extend the result of Dadush and Tiwari to show that Cutting Planes has short refutations of any unsatisfiable system of linear equations over a finite field. Like the Cutting Planes proofs of Dadush and Tiwari, our refutations also incur a quasi-polynomial blow-up in depth, and we conjecture that this is inherent. As a step towards this conjecture, we develop a new geometric technique for proving lower bounds on the depth of Cutting Planes proofs. This allows us to establish the first lower bounds on the depth of Semantic Cutting Planes proofs of the Tseitin formulas.

cs.CC↗

arXiv2021

Learning stochastic decision trees

We give a quasipolynomial-time algorithm for learning stochastic decision trees that is optimally resilient to adversarial noise. Given an $η$-corrupted set of uniform random samples labeled by a size-$s$ stochastic decision tree, our algorithm runs in time $n^{O(\log(s/\varepsilon)/\varepsilon^2)}$ and returns a hypothesis with error within an additive $2η+ \varepsilon$ of the Bayes optimal. An additive $2η$ is the information-theoretic minimum. Previously no non-trivial algorithm with a guarantee of $O(η) + \varepsilon$ was known, even for weaker noise models. Our algorithm is furthermore proper, returning a hypothesis that is itself a decision tree; previously no such algorithm was known even in the noiseless setting.

cs.LG↗

arXiv2020

Estimating decision tree learnability with polylogarithmic sample complexity

We show that top-down decision tree learning heuristics are amenable to highly efficient learnability estimation: for monotone target functions, the error of the decision tree hypothesis constructed by these heuristics can be estimated with polylogarithmically many labeled examples, exponentially smaller than the number necessary to run these heuristics, and indeed, exponentially smaller than information-theoretic minimum required to learn a good decision tree. This adds to a small but growing list of fundamental learning algorithms that have been shown to be amenable to learnability estimation. En route to this result, we design and analyze sample-efficient minibatch versions of top-down decision tree learning heuristics and show that they achieve the same provable guarantees as the full-batch versions. We further give "active local" versions of these heuristics: given a test point $x^\star$, we show how the label $T(x^\star)$ of the decision tree hypothesis $T$ can be computed with polylogarithmically many labeled examples, exponentially smaller than the number necessary to learn $T$.

cs.LG↗

arXiv2020

Query strategies for priced information, revisited

We consider the problem of designing query strategies for priced information, introduced by Charikar et al. In this problem the algorithm designer is given a function $f : \{0,1\}^n \to \{-1,1\}$ and a price associated with each of the $n$ coordinates. The goal is to design a query strategy for determining $f$'s value on unknown inputs for minimum cost. Prior works on this problem have focused on specific classes of functions. We analyze a simple and natural strategy that applies to all functions $f$, and show that its performance relative to the optimal strategy can be expressed in terms of a basic complexity measure of $f$, its influence. For $\varepsilon \in (0,\frac1{2})$, writing $\mathsf{opt}$ to denote the expected cost of the optimal strategy that errs on at most an $\varepsilon$-fraction of inputs, our strategy has expected cost $\mathsf{opt} \cdot \mathrm{Inf}(f)/\varepsilon^2$ and also errs on at most an $O(\varepsilon)$-fraction of inputs. This connection yields new guarantees that complement existing ones for a number of function classes that have been studied in this context, as well as new guarantees for new classes. Finally, we show that improving on the parameters that we achieve will require making progress on the longstanding open problem of properly learning decision trees.

cs.DS↗

arXiv2020

Universal guarantees for decision tree induction via a higher-order splitting criterion

We propose a simple extension of top-down decision tree learning heuristics such as ID3, C4.5, and CART. Our algorithm achieves provable guarantees for all target functions $f: \{-1,1\}^n \to \{-1,1\}$ with respect to the uniform distribution, circumventing impossibility results showing that existing heuristics fare poorly even for simple target functions. The crux of our extension is a new splitting criterion that takes into account the correlations between $f$ and small subsets of its attributes. The splitting criteria of existing heuristics (e.g. Gini impurity and information gain), in contrast, are based solely on the correlations between $f$ and its individual attributes. Our algorithm satisfies the following guarantee: for all target functions $f : \{-1,1\}^n \to \{-1,1\}$, sizes $s\in \mathbb{N}$, and error parameters $ε$, it constructs a decision tree of size $s^{\tilde{O}((\log s)^2/ε^2)}$ that achieves error $\le O(\mathsf{opt}_s) + ε$, where $\mathsf{opt}_s$ denotes the error of the optimal size $s$ decision tree. A key technical notion that drives our analysis is the noise stability of $f$, a well-studied smoothness measure.

cs.LG↗