arXiv subjects

Vaggos Chatziafratis

Publications and source records attributed to Vaggos Chatziafratis.

At least 19 recordsLinked to original sources

arXiv2026

Tunable Latent Generative Priors for Compressed Sensing and Inverse Problems

Latent generative models have emerged as powerful priors for solving inverse problems. These models typically represent a class of natural signals at a single, fixed complexity, governed by the latent dimensionality. This can be limiting: depending on the problem, a latent dimensionality that is too small may result in high representation error, while one that is too large may overfit to noise. We develop tunable latent priors for diffusion models, normalizing flows, and variational autoencoders, leveraging nested dropout. Across tasks including compressed sensing, inpainting, denoising, and phase retrieval, we show empirically that tunable priors consistently achieve lower reconstruction errors than fixed-complexity baselines. In the linear denoising setting, we derive the optimal complexity in closed form, showing how it depends on the noise level and the signal spectrum. This work demonstrates the potential of tunable latent generative priors and motivates both the development of supporting theory and their application across a wide range of inverse problems.

cs.LG↗

arXiv2026

Strong Refutation of Ordering, Phylogenetic, and Ordinary CSPs, and New Satisfiability and Refutation Thresholds for Triplet and Quartet Reconstruction

We study phase transitions and algorithms for refuting CSPs arising in hierarchical clustering (as well as ranking, and ordinary CSPs). Here, $n$ variables are assigned to leaves of a tree, so as to satisfy $m$ constraints, specifying evolutionary relationships. Two canonical $NP$-hard optimization problems are Triplet and Quartet Reconstruction, where the input consists of triplets $xy|z$ or quartets $xy|zw$, and the goal is to find a tree $T^*$ maximizing agreement with constraints. Our main results are (as density $λ=m/n$ increases): 1. We show the existence and precisely locate the sharp threshold $λ^*\approx1.2277$ for Triplets (via closed-form solution). To the best of our knowledge, this is the first sharp threshold for the broad family of Phylogenetic CSPs. Moreover, we give a lower and upper bound for Quartets. 2. We provide strong refutation algorithms that certify that $val(T^*)\le5/9 + ε$, where $val(T^*)$ is the fraction of constraints satisfied by the (unknown) optimal tree. For triplets, our algorithm succeeds w.h.p if $m =Ω(n)$, and for quartets if $m = Ω(n^{3/2})$. 3. We obtain strongest possible refutations at slightly larger densities (for triplets $m=O(n^{3/2}\log ^3n)$, for quartets $m=O(n^2)$): we certify that $T^*$ is no better than a random assignment, i.e., $val(T^*)\le 1/3+ε$. In fact, we obtain strongest possible refutations for finite-alphabet CSPs with or without negations. Our refutations above are instantiations of our general theorem that applies more broadly to Phylogenetic and Ordering CSPs (and all CSPs failing to support $t$-wise independence), and generalizes the current algorithmic frontier on refuting random CSPs~\citep{allen2015refute}. A crucial difference here, unlike Boolean CSPs, is that there are no negated variables, so prior works relying on negations -- a source of randomness -- do not apply.

cs.DS↗

arXiv2026

Provable Accuracy Collapse in Embedding-Based Representations under Dimensionality Mismatch

Embedding-based representations in Euclidean space $\mathbb{R}^d$ are a cornerstone of modern machine learning, where a major goal is to use the \emph{smallest dimension} that faithfully captures data relations. In this work, we prove sharp dimension--accuracy tradeoffs and identify a fundamental information-theoretic limitation: unless the embedding dimension $d$ is chosen close to the ground-truth dimension $D$, accuracy undergoes a sudden collapse. Our main result shows that this phenomenon arises even in standard contrastive learning settings, where supervision is limited to a set of $m$ anchor--positive--negative triplets $(i,j,k)$ encoding distance comparisons $\mathrm{dist}(i,j) < \mathrm{dist}(i,k)$. Specifically, given triplets realizable by an unknown ground-truth embedding in $D$ dimensions, we prove that there exists constant $c < 1$, such that \emph{every embedding of dimension at most $cD$ violates half of the triplets}, yielding accuracy as low as a trivial one-dimensional solution that ignores the input. We complement our information-theoretic bounds with strong computational hardness results: under the Unique Games Conjecture, even if the given triplets are nearly realizable in $D=1$ dimension, no polynomial-time algorithm -- \textit{regardless of its dimension} -- can achieve accuracy above the trivial $50\%$ baseline.

cs.DS↗

arXiv2026

Optimal Phylogenetic Reconstruction from Sampled Quartets

Quartet Reconstruction, the task of recovering a phylogenetic tree from smaller trees on four species called \textit{quartets}, is a well-studied problem in theoretical computer science with far-reaching connections to statistics, graph theory and biology. Given a random sample containing $m$ noisy quartets, labeled by an unknown ground-truth tree $T$ on $n$ taxa, we want to output a tree $\widehat T$ that is \textit{close} to $T$ in terms of quartet distance and can predict unseen quartets. Unfortunately, the empirical risk minimizer corresponds to the $\mathsf{NP}$-hard problem of finding a tree that maximizes agreements with the sampled quartets, and earlier works in approximation algorithms gave $(1-\eps)$-approximation schemes (PTAS) for dense instances with $m=Θ(n^4)$ quartets, or for $m=Θ(n^2\log n)$ quartets \textit{randomly} sampled from $T$. Prior to our work, it was unknown how many samples are information-theoretically required to learn the tree, and whether there is an efficient reconstruction algorithm. We present optimal results for reconstructing an unknown phylogenetic tree $T$ from a random sample of $m=Θ(n)$ quartets, corrupted under the Random Classification Noise (RCN) model. This matches the $Ω(n)$ lower bound required for any meaningful tree reconstruction. Our contribution is twofold: first, we give a tree reconstruction algorithm that, not only achieves a $(1-\eps)$-approximation, but most importantly \textit{recovers} a tree close to $T$ in quartet distance; second, we show a new $Θ(n)$ bound on the Natarajan dimension of phylogenies (an analog of VC dimension in multiclass classification). Our analysis relies on a new \textit{Quartet-based Embedding and Detection} procedure that identifies and removes well-clustered subtrees from the (unknown) ground-truth $T$ via semidefinite programming.

cs.DS↗

arXiv2025

Accelerating data-driven algorithm selection for combinatorial partitioning problems

Data-driven algorithm selection is a powerful approach for choosing effective heuristics for computational problems. It operates by evaluating a set of candidate algorithms on a collection of representative training instances and selecting the one with the best empirical performance. However, running each algorithm on every training instance is computationally expensive, making scalability a central challenge. In practice, a common workaround is to evaluate algorithms on smaller proxy instances derived from the original inputs. However, this practice has remained largely ad hoc and lacked theoretical grounding. We provide the first theoretical foundations for this practice by formalizing the notion of size generalization: predicting an algorithm's performance on a large instance by evaluating it on a smaller, representative instance, subsampled from the original instance. We provide size generalization guarantees for three widely used clustering algorithms (single-linkage, $k$-means++, and Gonzalez's $k$-centers heuristic) and two canonical max-cut algorithms (Goemans-Williamson and Greedy). We characterize the subsample size sufficient to ensure that performance on the subsample reflects performance on the full instance, and our experiments support these findings.

cs.LG↗

arXiv2025

The Complexity of Finding Local Optima in Contrastive Learning

Contrastive learning is a powerful technique for discovering meaningful data representations by optimizing objectives based on $\textit{contrastive information}$, often given as a set of weighted triplets $\{(x_i, y_i^+, z_{i}^-)\}_{i = 1}^m$ indicating that an "anchor" $x_i$ is more similar to a "positive" example $y_i$ than to a "negative" example $z_i$. The goal is to find representations (e.g., embeddings in $\mathbb{R}^d$ or a tree metric) where anchors are placed closer to positive than to negative examples. While finding $\textit{global}$ optima of contrastive objectives is $\mathsf{NP}$-hard, the complexity of finding $\textit{local}$ optima -- representations that do not improve by local search algorithms such as gradient-based methods -- remains open. Our work settles the complexity of finding local optima in various contrastive learning problems by proving $\mathsf{PLS}$-hardness in discrete settings (e.g., maximize satisfied triplets) and $\mathsf{CLS}$-hardness in continuous settings (e.g., minimize Triplet Loss), where $\mathsf{PLS}$ (Polynomial Local Search) and $\mathsf{CLS}$ (Continuous Local Search) are well-studied complexity classes capturing local search dynamics in discrete and continuous optimization, respectively. Our results imply that no polynomial time algorithm (local search or otherwise) can find a local optimum for various contrastive learning problems, unless $\mathsf{PLS}\subseteq\mathsf{P}$ (or $\mathsf{CLS}\subseteq \mathsf{P}$ for continuous problems). Even in the unlikely scenario that $\mathsf{PLS}\subseteq\mathsf{P}$ (or $\mathsf{CLS}\subseteq \mathsf{P}$), our reductions imply that there exist instances where local search algorithms need exponential time to reach a local optimum, even for $d=1$ (embeddings on a line).

cs.LG↗

arXiv2025

Welfare Approximation in Additively Separable Hedonic Games

Partitioning a set of $n$ items or agents while maximizing the value of the partition is a fundamental algorithmic task. We study this problem in the specific setting of maximizing social welfare in additively separable hedonic games. Unfortunately, this task faces strong computational boundaries: Extending previous results, we show that approximating welfare by a factor of $n^{1-ε}$ is NP-hard, even for severely restricted weights. However, we can obtain a randomized $\log n$-approximation on instances for which the sum of input valuations is nonnegative. Finally, we study two stochastic models of aversion-to-enemies games, where the weights are derived from Erdős-Rényi or multipartite graphs. We obtain constant-factor and logarithmic-factor approximations with high probability.

cs.GT↗

arXiv2023

Dimension-Accuracy Tradeoffs in Contrastive Embeddings for Triplets, Terminals & Top-k Nearest Neighbors

Metric embeddings traditionally study how to map $n$ items to a target metric space such that distance lengths are not heavily distorted; but what if we only care to preserve the relative order of the distances (and not their length)? In this paper, we are motivated by the following basic question: given triplet comparisons of the form ``item $i$ is closer to item $j$ than to item $k$,'' can we find low-dimensional Euclidean representations for the $n$ items that respect those distance comparisons? Such order-preserving embeddings naturally arise in important applications and have been studied since the 1950s, under the name of ordinal or non-metric embeddings. Our main results are: 1. Nearly-Tight Bounds on Triplet Dimension: We introduce the natural concept of triplet dimension of a dataset, and surprisingly, we show that in order for an ordinal embedding to be triplet-preserving, its dimension needs to grow as $\frac n2$ in the worst case. This is optimal (up to constant) as $n-1$ dimensions always suffice. 2. Tradeoffs for Dimension vs (Ordinal) Relaxation: We then relax the requirement that every triplet should be exactly preserved and present almost tight lower bounds for the maximum ratio between distances whose relative order was inverted by the embedding; this ratio is known as (ordinal) relaxation in the literature and serves as a counterpart to (metric) distortion. 3. New Bounds on Terminal and Top-$k$-NNs Embeddings: Going beyond triplets, we then study two well-motivated scenarios where we care about preserving specific sets of distances (not necessarily triplets). The first scenario is Terminal Ordinal Embeddings and the second scenario is top-$k$-NNs Ordinal Embeddings. To the best of our knowledge, these are some of the first tradeoffs on triplet-preserving ordinal embeddings and the first study of Terminal and Top-$k$-NNs Ordinal Embeddings.

cs.DS↗

arXiv2023

Triplet Reconstruction and all other Phylogenetic CSPs are Approximation Resistant

We study the natural problem of Triplet Reconstruction (also Rooted Triplets Consistency or Triplet Clustering), originally motivated in computational biology and relational databases (Aho, Sagiv, Szymanski, and Ullman, 1981): given $n$ points, we want to embed them onto the $n$ leaves of a rooted binary tree (a hierarchical clustering or ultrametric embedding) such that a given set of $m$ triplet constraints is satisfied. Triplet $ij|k$ indicates that ``$i, j$ are more closely related to each other than to $k$'' and a tree satisfies $ij|k$ if $d(i,j)$ is the smallest among the 3 distances. Aho et al. (1981) gave an elegant efficient algorithm to find a tree respecting all constraints (if it exists) and it is easy to see that a random binary tree is a 1/3-approximation. Unfortunately, despite more than four decades of research, no better approximation is known. Our main theorem--which captures Triplet Reconstruction as a special case--is a general hardness of approximation result about Constraint Satisfaction Problems (CSPs) over infinite domains (the variables are mapped to any of the $n$ leaves of a tree). Specifically, we prove, under Unique Games (Khot, 2002), that Triplet Reconstruction and more generally, every CSP over hierarchies is approximation resistant (there is no polynomial-time algorithm that does asymptotically better than a biased random assignment). This settles the approximability for many interesting Subtree or Supertree Aggregation Problems. More broadly, our result significantly extends the list of approximation resistant predicates and is a generalization of Guruswami, Hastad, Manokaran, Raghavendra, and Charikar (2011), who showed that ordering CSPs are approximation resistant. The main challenge in our analyses stems from the fact that trees have topology which is what determines whether a given triplet constraint on the leaves is satisfied or not.

cs.DS↗

arXiv2022

On Scrambling Phenomena for Randomly Initialized Recurrent Networks

Recurrent Neural Networks (RNNs) frequently exhibit complicated dynamics, and their sensitivity to the initialization process often renders them notoriously hard to train. Recent works have shed light on such phenomena analyzing when exploding or vanishing gradients may occur, either of which is detrimental for training dynamics. In this paper, we point to a formal connection between RNNs and chaotic dynamical systems and prove a qualitatively stronger phenomenon about RNNs than what exploding gradients seem to suggest. Our main result proves that under standard initialization (e.g., He, Xavier etc.), RNNs will exhibit \textit{Li-Yorke chaos} with \textit{constant} probability \textit{independent} of the network's width. This explains the experimentally observed phenomenon of \textit{scrambling}, under which trajectories of nearby points may appear to be arbitrarily close during some timesteps, yet will be far away in future timesteps. In stark contrast to their feedforward counterparts, we show that chaotic behavior in RNNs is preserved under small perturbations and that their expressive power remains exponential in the number of feedback iterations. Our technical arguments rely on viewing RNNs as random walks under non-linear activations, and studying the existence of certain types of higher-order fixed points called \textit{periodic points} that lead to phase transitions from order to chaos.

cs.LG↗

arXiv2022

Efficiently Computing Nash Equilibria in Adversarial Team Markov Games

Computing Nash equilibrium policies is a central problem in multi-agent reinforcement learning that has received extensive attention both in theory and in practice. However, provable guarantees have been thus far either limited to fully competitive or cooperative scenarios or impose strong assumptions that are difficult to meet in most practical applications. In this work, we depart from those prior results by investigating infinite-horizon \emph{adversarial team Markov games}, a natural and well-motivated class of games in which a team of identically-interested players -- in the absence of any explicit coordination or communication -- is competing against an adversarial player. This setting allows for a unifying treatment of zero-sum Markov games and Markov potential games, and serves as a step to model more realistic strategic interactions that feature both competing and cooperative interests. Our main contribution is the first algorithm for computing stationary $ε$-approximate Nash equilibria in adversarial team Markov games with computational complexity that is polynomial in all the natural parameters of the game, as well as $1/ε$. The proposed algorithm is particularly natural and practical, and it is based on performing independent policy gradient steps for each player in the team, in tandem with best responses from the side of the adversary; in turn, the policy for the adversary is then obtained by solving a carefully constructed linear program. Our analysis leverages non-standard techniques to establish the KKT optimality conditions for a nonlinear program with nonconvex constraints, thereby leading to a natural interpretation of the induced Lagrange multipliers. Along the way, we significantly extend an important characterization of optimal policies in adversarial (normal-form) team games due to Von Stengel and Koller (GEB `97).

cs.GT↗

arXiv2022

Hierarchical Clustering in Graph Streams: Single-Pass Algorithms and Space Lower Bounds

The Hierarchical Clustering (HC) problem consists of building a hierarchy of clusters to represent a given dataset. Motivated by the modern large-scale applications, we study the problem in the \streaming model, in which the memory is heavily limited and only a single or very few passes over the input are allowed. Specifically, we investigate whether a good hierarchical clustering can be obtained, or at least whether we can approximately estimate the value of the optimal hierarchy. To measure the quality of a hierarchy, we use the HC minimization objective introduced by Dasgupta. Assuming that the input is an $n$-vertex weighted graph whose edges arrive in a stream, we derive the following results on space-vs-accuracy tradeoffs: * With $O(n\cdot \text{polylog}\,{n})$ space, we develop a single-pass algorithm, whose approximation ratio matches the currently best offline algorithm. * When the space is more limited, namely, $n^{1-o(1)}$, we prove that no algorithm can even estimate the value of optimum HC tree to within an $o(\frac{\log{n}}{\log\log{n}})$ factor, even when allowed $\text{polylog}{\,{n}}$ passes over the input. * In the most stringent setting of $\text{polylog}\,{n}$ space, we rule out algorithms that can even distinguish between "highly"-vs-"poorly" clusterable graphs, namely, graphs that have an $n^{1/2-o(1)}$ factor gap between their HC objective value. * Finally, we prove that any single-pass streaming algorithm that computes an optimal HC tree requires to store almost the entire input even if allowed exponential time. Our algorithmic results establish a general structural result that proves that cut sparsifiers of input graph can preserve cost of "balanced" HC trees to within a constant factor. Our lower bound results include a new streaming lower bound for a novel problem "One-vs-Many-Expanders", which can be of independent interest.

cs.DS↗

arXiv2021

Bilu-Linial stability, certified algorithms and the Independent Set problem

We study the Maximum Independent Set (MIS) problem under the notion of stability introduced by Bilu and Linial (2010): a weighted instance of MIS is $γ$-stable if it has a unique optimal solution that remains the unique optimum under multiplicative perturbations of the weights by a factor of at most $γ\geq 1$. The goal then is to efficiently recover the unique optimal solution. In this work, we solve stable instances of MIS on several graphs classes: we solve $\widetilde{O}(Δ/\sqrt{\log Δ})$-stable instances on graphs of maximum degree $Δ$, $(k - 1)$-stable instances on $k$-colorable graphs and $(1 + \varepsilon)$-stable instances on planar graphs. For general graphs, we present a strong lower bound showing that there are no efficient algorithms for $O(n^{\frac{1}{2} - \varepsilon})$-stable instances of MIS, assuming the planted clique conjecture. We also give an algorithm for $(\varepsilon n)$-stable instances. As a by-product of our techniques, we give algorithms and lower bounds for stable instances of Node Multiway Cut. Furthermore, we prove a general result showing that the integrality gap of convex relaxations of several maximization problems reduces dramatically on stable instances. Moreover, we initiate the study of certified algorithms, a notion recently introduced by Makarychev and Makarychev (2018), which is a class of $γ$-approximation algorithms that satisfy one crucial property: the solution returned is optimal for a perturbation of the original instance. We obtain $Δ$-certified algorithms for MIS on graphs of maximum degree $Δ$, and $(1+\varepsilon)$-certified algorithms on planar graphs. Finally, we analyze the algorithm of Berman and Furer (1994) and prove that it is a $\left(\frac{Δ+ 1}{3} + \varepsilon\right)$-certified algorithm for MIS on graphs of maximum degree $Δ$ where all weights are equal to 1.

cs.DS↗

arXiv2021

Expressivity of Neural Networks via Chaotic Itineraries beyond Sharkovsky's Theorem

Given a target function $f$, how large must a neural network be in order to approximate $f$? Recent works examine this basic question on neural network \textit{expressivity} from the lens of dynamical systems and provide novel ``depth-vs-width'' tradeoffs for a large family of functions $f$. They suggest that such tradeoffs are governed by the existence of \textit{periodic} points or \emph{cycles} in $f$. Our work, by further deploying dynamical systems concepts, illuminates a more subtle connection between periodicity and expressivity: we prove that periodic points alone lead to suboptimal depth-width tradeoffs and we improve upon them by demonstrating that certain ``chaotic itineraries'' give stronger exponential tradeoffs, even in regimes where previous analyses only imply polynomial gaps. Contrary to prior works, our bounds are nearly-optimal, tighten as the period increases, and handle strong notions of inapproximability (e.g., constant $L_1$ error). More broadly, we identify a phase transition to the \textit{chaotic regime} that exactly coincides with an abrupt shift in other notions of function complexity, including VC-dimension and topological entropy.

cs.LG↗

arXiv2021

Adversarially Robust Low Dimensional Representations

Many machine learning systems are vulnerable to small perturbations made to inputs either at test time or at training time. This has received much recent interest on the empirical front due to applications where reliability and security are critical. However, theoretical understanding of algorithms that are robust to adversarial perturbations is limited. In this work we focus on Principal Component Analysis (PCA), a ubiquitous algorithmic primitive in machine learning. We formulate a natural robust variant of PCA where the goal is to find a low dimensional subspace to represent the given data with minimum projection error, that is in addition robust to small perturbations measured in $\ell_q$ norm (say $q=\infty$). Unlike PCA which is solvable in polynomial time, our formulation is computationally intractable to optimize as it captures a variant of the well-studied sparse PCA objective as a special case. We show the following results: -Polynomial time algorithm that is constant factor competitive in the worst-case with respect to the best subspace, in terms of the projection error and the robustness criterion. -We show that our algorithmic techniques can also be made robust to adversarial training-time perturbations, in addition to yielding representations that are robust to adversarial perturbations at test time. Specifically, we design algorithms for a strong notion of training-time perturbations, where every point is adversarially perturbed up to a specified amount. -We illustrate the broad applicability of our algorithmic techniques in addressing robustness to adversarial perturbations, both at training time and test time. In particular, our adversarially robust PCA primitive leads to computationally efficient and robust algorithms for both unsupervised and supervised learning problems such as clustering and learning adversarially robust classifiers.

cs.DS↗

arXiv2021

Maximizing Agreements for Ranking, Clustering and Hierarchical Clustering via MAX-CUT

In this paper, we study a number of well-known combinatorial optimization problems that fit in the following paradigm: the input is a collection of (potentially inconsistent) local relationships between the elements of a ground set (e.g., pairwise comparisons, similar/dissimilar pairs, or ancestry structure of triples of points), and the goal is to aggregate this information into a global structure (e.g., a ranking, a clustering, or a hierarchical clustering) in a way that maximizes agreement with the input. Well-studied problems such as rank aggregation, correlation clustering, and hierarchical clustering with triplet constraints fall in this class of problems. We study these problems on stochastic instances with a hidden embedded ground truth solution. Our main algorithmic contribution is a unified technique that uses the maximum cut problem in graphs to approximately solve these problems. Using this technique, we can often get approximation guarantees in the stochastic setting that are better than the known worst case inapproximability bounds for the corresponding problem. On the negative side, we improve the worst case inapproximability bound on several hierarchical clustering formulations through a reduction to related ranking problems.

cs.DS↗

arXiv2021

Hierarchical Clustering via Sketches and Hierarchical Correlation Clustering

Recently, Hierarchical Clustering (HC) has been considered through the lens of optimization. In particular, two maximization objectives have been defined. Moseley and Wang defined the \emph{Revenue} objective to handle similarity information given by a weighted graph on the data points (w.l.o.g., $[0,1]$ weights), while Cohen-Addad et al. defined the \emph{Dissimilarity} objective to handle dissimilarity information. In this paper, we prove structural lemmas for both objectives allowing us to convert any HC tree to a tree with constant number of internal nodes while incurring an arbitrarily small loss in each objective. Although the best-known approximations are 0.585 and 0.667 respectively, using our lemmas we obtain approximations arbitrarily close to 1, if not all weights are small (i.e., there exist constants $ε, δ$ such that the fraction of weights smaller than $δ$, is at most $1 - ε$); such instances encompass many metric-based similarity instances, thereby improving upon prior work. Finally, we introduce Hierarchical Correlation Clustering (HCC) to handle instances that contain similarity and dissimilarity information simultaneously. For HCC, we provide an approximation of 0.4767 and for complementary similarity/dissimilarity weights (analogous to $+/-$ correlation clustering), we again present nearly-optimal approximations.

cs.DS↗

arXiv2020

Inapproximability for Local Correlation Clustering and Dissimilarity Hierarchical Clustering

We present hardness of approximation results for Correlation Clustering with local objectives and for Hierarchical Clustering with dissimilarity information. For the former, we study the local objective of Puleo and Milenkovic (ICML '16) that prioritizes reducing the disagreements at data points that are worst off and for the latter we study the maximization version of Dasgupta's cost function (STOC '16). Our APX hardness results imply that the two problems are hard to approximate within a constant of 4/3 ~ 1.33 (assuming P vs NP) and 9159/9189 ~ 0.9967 (assuming the Unique Games Conjecture) respectively.

cs.DS↗