arXiv subjects

Daniel Stefankovic

Publications and source records attributed to Daniel Stefankovic.

At least 19 recordsLinked to original sources

arXiv2026

Critical window for approximate counting in dense Ising models

We study the complexity of approximating the partition function of dense Ising models in the critical regime. Recent work of Chen, Chen, Yin, and Zhang (FOCS 2025) established fast mixing at criticality, and even beyond criticality in a window of width $N^{-1/2}$. We complement these algorithmic results by proving nearly tight hardness bounds, thus yielding the first instance of a sharp scaling window for the computational complexity of approximate counting. Specifically, for the dense Ising model we show that approximating the partition function is computationally hard within a window of width $N^{-1/2+\varepsilon}$ for any constant $\varepsilon>0$. Standard hardness reductions for non-critical regimes break down at criticality due to bigger fluctuations in the underlying gadgets, leading to suboptimal bounds. We overcome this barrier via a global approach which aggregates fluctuations across all gadgets rather than requiring tight concentration guarantees for each individually. This new approach yields the optimal exponent for the critical window.

cs.CC↗

arXiv2023

Spectral Independence and Local-to-Global Techniques for Optimal Mixing of Markov Chains

This monograph is an exposition on an exciting new technique known as spectral independence, which has been instrumental in analyzing the convergence rate of Markov Chain Monte Carlo (MCMC) algorithms. For a high-dimensional distribution defined on labelings of the vertices of an $n$-vertex graph, the spectral independence condition, introduced by Anari, Liu, and Oveis Gharan (2020), is a bound on the maximum eigenvalue of the influence matrix capturing the influence between pairs of vertices (closely related to the covariance between the variables). In the first part of the monograph, we present results showing that spectral independence (and related techniques) imply fast mixing of simple Markov chains such as the Glauber dynamics (aka Gibbs sampler). These proofs rely on local-to-global theorems relating local walks encoding pairwise correlations to variance decay and mixing properties of Markov chains. We focus on two applications: the hard-core model on independent sets of a graph (which is a combinatorial example of a binary graphical model) and random bases of a matroid. We apply the techniques presented in this monograph to show recent results of fast mixing of the Glauber dynamics on general graphs in the so-called tree-uniqueness region, polynomial-time mixing on general graphs at the critical point for the uniqueness threshold, polynomial-time mixing on random regular graphs beyond the uniqueness threshold, and fast mixing of the bases-exchange walk for generating a random basis of an arbitrary matroid. Our focus in this monograph is on the analysis of the spectral gap of the associated Markov chains from a functional analysis perspective; we present proofs of the associated local-to-global theorems and the Trickle-Down Theorem from this same Markov chain perspective. The monograph is self-contained and aims to present the proofs in a unified fashion.

cs.DM↗

arXiv2023

Optimal Mixing via Tensorization for Random Independent Sets on Arbitrary Trees

We study the mixing time of the single-site update Markov chain, known as the Glauber dynamics, for generating a random independent set of a tree. Our focus is obtaining optimal convergence results for arbitrary trees. We consider the more general problem of sampling from the Gibbs distribution in the hard-core model where independent sets are weighted by a parameter $\lambda>0$; the special case $\lambda=1$ corresponds to the uniform distribution over all independent sets. Previous work of Martinelli, Sinclair and Weitz (2004) obtained optimal mixing time bounds for the complete $\Delta$-regular tree for all $\lambda$. However, Restrepo et al. (2014) showed that for sufficiently large $\lambda$ there are bounded-degree trees where optimal mixing does not hold. Recent work of Eppstein and Frishberg (2022) proved a polynomial mixing time bound for the Glauber dynamics for arbitrary trees, and more generally for graphs of bounded tree-width. We establish an optimal bound on the relaxation time (i.e., inverse spectral gap) of $O(n)$ for the Glauber dynamics for unweighted independent sets on arbitrary trees. We stress that our results hold for arbitrary trees and there is no dependence on the maximum degree $\Delta$. Interestingly, our results extend (far) beyond the uniqueness threshold which is on the order $\lambda=O(1/\Delta)$. Our proof approach is inspired by recent work on spectral independence. In fact, we prove that spectral independence holds with a constant independent of the maximum degree for any tree, but this does not imply mixing for general trees as the optimal mixing results of Chen, Liu, and Vigoda (2021) only apply for bounded degree graphs. We instead utilize the combinatorial nature of independent sets to directly prove approximate tensorization of variance via a non-trivial inductive proof.

cs.DM↗

arXiv2022

Beyond the Existential Theory of the Reals

We show that completeness at higher levels of the theory of the reals is a robust notion (under changing the signature and bounding the domain of the quantifiers). This mends recognized gaps in the hierarchy, and leads to stronger completeness results for various computational problems. We exhibit several families of complete problems which can be used for future completeness results in the real hierarchy. As an application we sharpen some results by B\"{u}rgisser and Cucker on the complexity of properties of semialgebraic sets, including the Hausdorff distance problem also studied by Jungeblut, Kleist, and Miltzow.

cs.CC↗

arXiv2022

Approximating observables is as hard as counting

We study the computational complexity of estimating local observables for Gibbs distributions. A simple combinatorial example is the average size of an independent set in a graph. In a recent work, we established NP-hardness of approximating the average size of an independent set utilizing hardness of the corresponding optimization problem and the related phase transition behavior. Here, we instead consider settings where the underlying optimization problem is easily solvable. Our main contribution is to classify the complexity of approximating a wide class of observables via a generic reduction from approximate counting to the problem of estimating local observables. The key idea is to use the observables to interpolate the counting problem. Using this new approach, we are able to study observables on bipartite graphs where the underlying optimization problem is easy but the counting problem is believed to be hard. The most-well studied class of graphs that was excluded from previous hardness results were bipartite graphs. We establish hardness for estimating the average size of the independent set in bipartite graphs of maximum degree 6; more generally, we show tight hardness results for general vertex-edge observables for antiferromagnetic 2-spin systems on bipartite graphs. Our techniques go beyond 2-spin systems, and for the ferromagnetic Potts model we establish hardness of approximating the number of monochromatic edges in the same region as known hardness of approximate counting results.

cs.CC↗

arXiv2022

Metastability of the Potts ferromagnet on random regular graphs

We study the performance of Markov chains for the $q$-state ferromagnetic Potts model on random regular graphs. It is conjectured that their performance is dictated by metastability phenomena, i.e., the presence of "phases" (clusters) in the sample space where Markov chains with local update rules, such as the Glauber dynamics, are bound to take exponential time to escape. The phases that are believed to drive these metastability phenomena in the case of the Potts model emerge as local, rather than global, maxima of the so-called Bethe functional, and previous approaches of analysing these phases based on optimisation arguments fall short of the task. Our first contribution is to detail the emergence of the metastable phases for the $q$-state Potts model on the $d$-regular random graph for all integers $q,d\geq 3$, and establish that for an interval of temperatures, which is delineated by the uniqueness and a broadcasting threshold on the $d$-regular tree, the two phases coexist. The proofs are based on a conceptual connection between spatial properties and the structure of the Potts distribution on the random regular graph, rather than complicated moment calculations. Based on this new structural understanding of the model, we obtain various algorithmic consequences. We first complement recent fast mixing results for Glauber dynamics by Blanca and Gheissari below the uniqueness threshold, showing an exponential lower bound on the mixing time above the uniqueness threshold. Then, we obtain tight results even for the non-local Swendsen-Wang chain, where we establish slow mixing/metastability for the whole interval of temperatures where the chain is conjectured to mix slowly on the random regular graph. The key is to bound the conductance of the chains using a random graph "planting" argument combined with delicate bounds on random-graph percolation.

math.PR↗

arXiv2020

Statistical Mechanical Analysis of Neural Network Pruning

Deep learning architectures with a huge number of parameters are often compressed using pruning techniques to ensure computational efficiency of inference during deployment. Despite multitude of empirical advances, there is a lack of theoretical understanding of the effectiveness of different pruning methods. We inspect different pruning techniques under the statistical mechanics formulation of a teacher-student framework and derive their generalization error (GE) bounds. It has been shown that Determinantal Point Process (DPP) based node pruning method is notably superior to competing approaches when tested on real datasets. Using GE bounds in the aforementioned setup we provide theoretical guarantees for their empirical observations. Another consistent finding in literature is that sparse neural networks (edge pruned) generalize better than dense neural networks (node pruned) for a fixed number of parameters. We use our theoretical setup to prove this finding and show that even the baseline random edge pruning method performs better than the DPP node pruning method. We also validate this empirically on real datasets.

cs.LG↗

arXiv2020

The complexity of approximating averages on bounded-degree graphs

We prove that, unless P=NP, there is no polynomial-time algorithm to approximate within some multiplicative constant the average size of an independent set in graphs of maximum degree 6. This is a special case of a more general result for the hard-core model defined on independent sets weighted by a parameter $\lambda>0$. In the general setting, we prove that, unless P=NP, for all $\Delta\geq 3$, all $\lambda>\lambda_c(\Delta)$, there is no FPTAS which applies to all graphs of maximum degree $\Delta$ for computing the average size of the independent set in the Gibbs distribution, where $\lambda_c(\Delta)$ is the critical point for the uniqueness/non-uniqueness phase transition on the $\Delta$-regular tree. Moreover, we prove that for $\lambda$ in a dense set of this non-uniqueness region the problem is NP-hard to approximate within some constant factor. Our work extends to the antiferromagnetic Ising model and generalizes to all 2-spin antiferromagnetic models, establishing hardness of computing the average magnetization in the tree non-uniqueness region. Previously, Schulman, Sinclair and Srivastava (2015) showed that it is #P-hard to compute the average magnetization exactly, but no hardness of approximation results were known. Hardness results of Sly (2010) and Sly and Sun (2014) for approximating the partition function do not imply hardness of computing averages. The new ingredient in our reduction is an intricate construction of pairs of rooted trees whose marginal distributions at the root agree but their derivatives disagree. The main technical contribution is controlling what marginal distributions and derivatives are achievable and using Cauchy's functional equation to argue existence of the gadgets.

cs.CC↗

arXiv2019

Sampling in Uniqueness from the Potts and Random-Cluster Models on Random Regular Graphs

We consider the problem of sampling from the Potts model on random regular graphs. It is conjectured that sampling is possible when the temperature of the model is in the uniqueness regime of the regular tree, but positive algorithmic results have been for the most part elusive. In this paper, for all integers $q\geq 3$ and $Δ\geq 3$, we develop algorithms that produce samples within error $o(1)$ from the $q$-state Potts model on random $Δ$-regular graphs, whenever the temperature is in uniqueness, for both the ferromagnetic and antiferromagnetic cases. The algorithm for the antiferromagnetic Potts model is based on iteratively adding the edges of the graph and resampling a bichromatic class that contains the endpoints of the newly added edge. Key to the algorithm is how to perform the resampling step efficiently since bichromatic classes may induce linear-sized components. To this end, we exploit the tree uniqueness to show that the average growth of bichromatic components is typically small, which allows us to use correlation decay algorithms for the resampling step. While the precise uniqueness threshold on the tree is not known for general values of $q$ and $Δ$ in the antiferromagnetic case, our algorithm works throughout uniqueness regardless of its value. In the case of the ferromagnetic Potts model, we simplify the algorithm significantly by utilising the random-cluster representation of the model. In particular, we show that a percolation-type algorithm succeeds in sampling from the random-cluster model with parameters $p,q$ on random $Δ$-regular graphs for all values of $q\geq 1$ and $p<p_c(q,Δ)$, where $p_c(q,Δ)$ corresponds to a uniqueness threshold for the model on the $Δ$-regular tree. When restricted to integer values of $q$, this yields a simplified algorithm for the ferromagnetic Potts model on random $Δ$-regular graphs.

cs.DM↗

arXiv2019

Improved Strong Spatial Mixing for Colorings on Trees

Strong spatial mixing (SSM) is a form of correlation decay that has played an essential role in the design of approximate counting algorithms for spin systems. A notable example is the algorithm of Weitz (2006) for the hard-core model on weighted independent sets. We study SSM for the $q$-colorings problem on the infinite $(d+1)$-regular tree. Weak spatial mixing (WSM) captures whether the influence of the leaves on the root vanishes as the height of the tree grows. Jonasson (2002) established WSM when $q>d+1$. In contrast, in SSM, we first fix a coloring on a subset of internal vertices, and we again ask if the influence of the leaves on the root is vanishing. It was known that SSM holds on the $(d+1)$-regular tree when $q>αd$ where $α\approx 1.763...$ is a constant that has arisen in a variety of results concerning random colorings. Here we improve on this bound by showing SSM for $q>1.59d$. Our proof establishes an $L^2$ contraction for the BP operator. For the contraction we bound the norm of the BP Jacobian by exploiting combinatorial properties of the coloring of the tree.

cs.DM↗

arXiv2019

Approximation via Correlation Decay when Strong Spatial Mixing Fails

Approximate counting via correlation decay is the core algorithmic technique used in the sharp delineation of the computational phase transition that arises in the approximation of the partition function of anti-ferromagnetic two-spin models. Previous analyses of correlation-decay algorithms implicitly depended on the occurrence of strong spatial mixing (SSM). This means that one uses worst-case analysis of the recursive procedure that creates the sub-instances. We develop a new analysis method that is more refined than the worst-case analysis. We take the shape of instances in the computation tree into consideration and amortise against certain "bad" instances that are created as the recursion proceeds. This enables us to show correlation decay and to obtain an FPTAS even when SSM fails. We apply our technique to the problem of approximately counting independent sets in hypergraphs with degree upper-bound Delta and with a lower bound k on the arity of hyperedges. Liu and Lin gave an FPTAS for k>=2 and Delta<=5 (lack of SSM was the obstacle preventing this algorithm from being generalised to Delta=6). Our technique gives a tight result for Delta=6, showing that there is an FPTAS for k>=3 and Delta<=6. The best previously-known approximation scheme for Delta=6 is the Markov-chain simulation based FPRAS of Bordewich, Dyer and Karpinski, which only works for k>=8. Our technique also applies for larger values of k, giving an FPTAS for k>=Delta. This bound is not substantially stronger than existing randomised results in the literature. Nevertheless, it gives the first deterministic approximation scheme in this regime. Moreover, unlike existing results, it leads to an FPTAS for counting dominating sets in regular graphs with sufficiently large degree. We further demonstrate that approximately counting independent sets in hypergraphs is NP-hard even within the uniqueness regime.

cs.CC↗

arXiv2018

The complexity of approximating the matching polynomial in the complex plane

We study the problem of approximating the value of the matching polynomial on graphs with edge parameter $\gamma$, where $\gamma$ takes arbitrary values in the complex plane. When $\gamma$ is a positive real, Jerrum and Sinclair showed that the problem admits an FPRAS on general graphs. For general complex values of $\gamma$, Patel and Regts, building on methods developed by Barvinok, showed that the problem admits an FPTAS on graphs of maximum degree $\Delta$ as long as $\gamma$ is not a negative real number less than or equal to $-1/(4(\Delta-1))$. Our first main result completes the picture for the approximability of the matching polynomial on bounded degree graphs. We show that for all $\Delta\geq 3$ and all real $\gamma$ less than $-1/(4(\Delta-1))$, the problem of approximating the value of the matching polynomial on graphs of maximum degree $\Delta$ with edge parameter $\gamma$ is #P-hard. We then explore whether the maximum degree parameter can be replaced by the connective constant. Sinclair et al. showed that for positive real $\gamma$ it is possible to approximate the value of the matching polynomial using a correlation decay algorithm on graphs with bounded connective constant (and potentially unbounded maximum degree). We first show that this result does not extend in general in the complex plane; in particular, the problem is #P-hard on graphs with bounded connective constant for a dense set of $\gamma$ values on the negative real axis. Nevertheless, we show that the result does extend for any complex value $\gamma$ that does not lie on the negative real axis. Our analysis accounts for complex values of $\gamma$ using geodesic distances in the complex plane in the metric defined by an appropriate density function.

cs.DM↗

arXiv2017

Swendsen-Wang Algorithm on the Mean-Field Potts Model

We study the $q$-state ferromagnetic Potts model on the $n$-vertex complete graph known as the mean-field (Curie-Weiss) model. We analyze the Swendsen-Wang algorithm which is a Markov chain that utilizes the random cluster representation for the ferromagnetic Potts model to recolor large sets of vertices in one step and potentially overcomes obstacles that inhibit single-site Glauber dynamics. Long et al. studied the case $q=2$, the Swendsen-Wang algorithm for the mean-field ferromagnetic Ising model, and showed that the mixing time satisfies: (i) $Θ(1)$ for $β<β_c$, (ii) $Θ(n^{1/4})$ for $β=β_c$, (iii) $Θ(\log n)$ for $β>β_c$, where $β_c$ is the critical temperature for the ordered/disordered phase transition. In contrast, for $q\geq 3$ there are two critical temperatures $0<β_u<β_{rc}$ that are relevant. We prove that the mixing time of the Swendsen-Wang algorithm for the ferromagnetic Potts model on the $n$-vertex complete graph satisfies: (i) $Θ(1)$ for $β<β_u$, (ii) $Θ(n^{1/3})$ for $β=β_u$, (iii) $\exp(n^{Ω(1)})$ for $β_u<β<β_{rc}$, and (iv) $Θ(\log{n})$ for $β\geqβ_{rc}$. These results complement refined results of Cuff et al. on the mixing time of the Glauber dynamics for the ferromagnetic Potts model.

cs.DM↗

arXiv2017

Inapproximability of the independent set polynomial in the complex plane

We study the complexity of approximating the independent set polynomial $Z_G(\lambda)$ of a graph $G$ with maximum degree $\Delta$ when the activity $\lambda$ is a complex number. This problem is already well understood when $\lambda$ is real using connections to the $\Delta$-regular tree $T$. The key concept in that case is the "occupation ratio" of the tree $T$. This ratio is the contribution to $Z_T(\lambda)$ from independent sets containing the root of the tree, divided by $Z_T(\lambda)$ itself. If $\lambda$ is such that the occupation ratio converges to a limit, as the height of $T$ grows, then there is an FPTAS for approximating $Z_G(\lambda)$ on a graph $G$ with maximum degree $\Delta$. Otherwise, the approximation problem is NP-hard. Unsurprisingly, the case where $\lambda$ is complex is more challenging. Peters and Regts identified the complex values of $\lambda$ for which the occupation ratio of the $\Delta$-regular tree converges. These values carve a cardioid-shaped region $\Lambda_\Delta$ in the complex plane. Motivated by the picture in the real case, they asked whether $\Lambda_\Delta$ marks the true approximability threshold for general complex values $\lambda$. Our main result shows that for every $\lambda$ outside of $\Lambda_\Delta$, the problem of approximating $Z_G(\lambda)$ on graphs $G$ with maximum degree at most $\Delta$ is indeed NP-hard. In fact, when $\lambda$ is outside of $\Lambda_\Delta$ and is not a positive real number, we give the stronger result that approximating $Z_G(\lambda)$ is actually #P-hard. If $\lambda$ is a negative real number outside of $\Lambda_\Delta$, we show that it is #P-hard to even decide whether $Z_G(\lambda)>0$, resolving in the affirmative a conjecture of Harvey, Srivastava and Vondrak. Our proof techniques are based around tools from complex analysis -- specifically the study of iterative multivariate rational maps.

cs.CC↗

arXiv2017

Sampling Random Colorings of Sparse Random Graphs

We study the mixing properties of the single-site Markov chain known as the Glauber dynamics for sampling $k$-colorings of a sparse random graph $G(n,d/n)$ for constant $d$. The best known rapid mixing results for general graphs are in terms of the maximum degree $Δ$ of the input graph $G$ and hold when $k>11Δ/6$ for all $G$. Improved results hold when $k>αΔ$ for graphs with girth $\geq 5$ and $Δ$ sufficiently large where $α\approx 1.7632\ldots$ is the root of $α=\exp(1/α)$; further improvements on the constant $α$ hold with stronger girth and maximum degree assumptions. For sparse random graphs the maximum degree is a function of $n$ and the goal is to obtain results in terms of the expected degree $d$. The following rapid mixing results for $G(n,d/n)$ hold with high probability over the choice of the random graph for sufficiently large constant~$d$. Mossel and Sly (2009) proved rapid mixing for constant $k$, and Efthymiou (2014) improved this to $k$ linear in~$d$. The condition was improved to $k>3d$ by Yin and Zhang (2016) using non-MCMC methods. Here we prove rapid mixing when $k>αd$ where $α\approx 1.7632\ldots$ is the same constant as above. Moreover we obtain $O(n^{3})$ mixing time of the Glauber dynamics, while in previous rapid mixing results the exponent was an increasing function in $d$. As in previous results for random graphs our proof analyzes an appropriately defined block dynamics to "hide" high-degree vertices. One new aspect in our improved approach is utilizing so-called local uniformity properties for the analysis of block dynamics. To analyze the "burn-in" phase we prove a concentration inequality for the number of disagreements propagating in large blocks.

cs.DM↗

arXiv2017

On The Projection Operator to A Three-view Cardinality Constrained Set

The cardinality constraint is an intrinsic way to restrict the solution structure in many domains, for example, sparse learning, feature selection, and compressed sensing. To solve a cardinality constrained problem, the key challenge is to solve the projection onto the cardinality constraint set, which is NP-hard in general when there exist multiple overlapped cardinality constraints. In this paper, we consider the scenario where the overlapped cardinality constraints satisfy a Three-view Cardinality Structure (TVCS), which reflects the natural restriction in many applications, such as identification of gene regulatory networks and task-worker assignment problem. We cast the projection into a linear programming, and show that for TVCS, the vertex solution of this linear programming is the solution for the original projection problem. We further prove that such solution can be found with the complexity proportional to the number of variables and constraints. We finally use synthetic experiments and two interesting applications in bioinformatics and crowdsourcing to validate the proposed TVCS model and method.

cs.LG↗

arXiv2017

Rapid Mixing Swendsen-Wang Sampler for Stochastic Partitioned Attractive Models

The Gibbs sampler is a particularly popular Markov chain used for learning and inference problems in Graphical Models (GMs). These tasks are computationally intractable in general, and the Gibbs sampler often suffers from slow mixing. In this paper, we study the Swendsen-Wang dynamics which is a more sophisticated Markov chain designed to overcome bottlenecks that impede the Gibbs sampler. We prove O(\log n) mixing time for attractive binary pairwise GMs (i.e., ferromagnetic Ising models) on stochastic partitioned graphs having n vertices, under some mild conditions, including low temperature regions where the Gibbs sampler provably mixes exponentially slow. Our experiments also confirm that the Swendsen-Wang sampler significantly outperforms the Gibbs sampler when they are used for learning parameters of attractive GMs.

cs.LG↗

arXiv2016

Implementations and the independent set polynomial below the Shearer threshold

The independent set polynomial is important in many areas. For every integer $\Delta\geq 2$, the Shearer threshold is the value $\lambda^*(\Delta)=(\Delta-1)^{\Delta-1}/\Delta^{\Delta}$ . It is known that for $\lambda < - \lambda^*(\Delta)$, there are graphs~$G$ with maximum degree~$\Delta$ whose independent set polynomial, evaluated at~$\lambda$, is at most~$0$. Also, there are no such graphs for any $\lambda > -\lambda^*(\Delta)$. This paper is motivated by the computational problem of approximating the independent set polynomial when $\lambda < - \lambda^*(\Delta)$. The key issue in complexity bounds for this problem is "implementation". Informally, an implementation of a real number $\lambda'$ is a graph whose hard-core partition function, evaluated at~$\lambda$, simulates a vertex-weight of~$\lambda'$ in the sense that $\lambda'$ is the ratio between the contribution to the partition function from independent sets containing a certain vertex and the contribution from independent sets that do not contain that vertex. Implementations are the cornerstone of intractability results for the problem of approximately evaluating the independent set polynomial. Our main result is that, for any $\lambda < - \lambda^*(\Delta)$, it is possible to implement a set of values that is dense over the reals. The result is tight in the sense that it is not possible to implement a set of values that is dense over the reals for any $\lambda> \lambda^*(\Delta)$. Our result has already been used in a paper with \bezakova{} (STOC 2018) to show that it is \#P-hard to approximate the evaluation of the independent set polynomial on graphs of degree at most~$\Delta$ at any value $\lambda<-\lambda^*(\Delta)$. In the appendix, we give an additional incomparable inapproximability result (strengthening the inapproximability bound to an exponential factor, but weakening the hardness to NP-hardness).

cs.CC↗