arXiv research

New approach for testable learning using moment matching and Rademacher complexity.

Polynomial-time tester-learner for general halfspaces with Gaussian adversarial noise.

Polynomial-time algorithm for learning halfspaces with Gaussian-distributed data and adversarial noise.

A new calibration metric bridges testability and actionability.

Understanding statistical inference under possibly non-sparse high-dimensional models has gained much interest recently. For a given component of the regression coefficient, we show that the difficulty of the problem depends on the sparsity of the corresponding row of the precision matrix of the covariates, not the spa…

New quantum codes improve error correction with local tests.

Efficient algorithm for learning halfspaces in a new model with polynomial time complexity.

New calibration measure SCDL improves trust in AI predictions.

Significant pattern mining, the problem of finding itemsets that are significantly enriched in one class of objects, is statistically challenging, as the large space of candidate patterns leads to an enormous multiple testing problem. Recently, the concept of testability was proposed as one approach to correct for mult…

Local learning method selects covariates for causal effect estimation in the presence of latent variables.

New theory for local parameterization of deep ReLU networks.

Reinforcement learning (RL) tasks are challenging to implement, execute and test due to algorithmic instability, hyper-parameter sensitivity, and heterogeneous distributed communication patterns. We argue for the separation of logical component composition, backend graph definition, and distributed execution. To this e…

Universal tester-learner for halfspaces over structured distributions.

The paper introduces tests for missing data models based on graph assumptions.

Representation and learning of long-range dependencies is a central challenge confronted in modern applications of machine learning to sequence data. Yet despite the prominence of this issue, the basic problem of measuring long-range dependence, either in a given data source or as represented in a trained deep model, r…

Probabilistic graphical models (PGMs) have become a popular tool for computational analysis of biological data in a variety of domains. But, what exactly are they and how do they work? How can we use PGMs to discover patterns that are biologically relevant? And to what extent can PGMs help us formulate new hypotheses t…

We study random 2-dimensional complexes in the Linial - Meshulam model and find torsion in their fundamental groups at various regimes. We find a simple algorithmically testable criterion for a subcomplex of a random 2-complex to be aspherical; this implies that any aspherical subcomplex of a random 2-complex satisfies…

As machine learning algorithms are increasingly applied to high impact yet high risk tasks, such as medical diagnosis or autonomous driving, it is critical that researchers can explain how such algorithms arrived at their predictions. In recent years, a number of image saliency methods have been developed to summarize …

Torch-Struct simplifies structured prediction for deep learning.

Generative model solves financial market equilibria with stable reinforcement learning.

Traditional anatomical analyses captured only a fraction of real phenomic information. Here, we apply deep learning to quantify total phenotypic similarity across 2468 butterfly photographs, covering 38 subspecies from the polymorphic mimicry complex of $\textit{Heliconius erato}$ and $\textit{Heliconius melpomene}$. E…

The semantic map calibrates uncertainty from language model probabilities.

A new approach to learning in brain-like networks using adversarial algorithms.

Finding statistically significant interactions between binary variables is computationally and statistically challenging in high-dimensional settings, due to the combinatorial explosion in the number of hypotheses. Terada et al. recently showed how to elegantly address this multiple testing problem by excluding non-tes…

The paper clarifies the approximation of SGD with Ito SDEs for finite learning rates.

We propose a method to learn causal response representations through direct effect analysis.

New technologies for recording the activity of large neural populations during complex behavior provide exciting opportunities for investigating the neural computations that underlie perception, cognition, and decision-making. Nonlinear state space models provide an interpretable signal processing framework by combinin…

Synthesizes computational approaches to understand neural timescales.

Increasing the batch size is a popular way to speed up neural network training, but beyond some critical batch size, larger batch sizes yield diminishing returns. In this work, we study how the critical batch size changes based on properties of the optimization algorithm, including acceleration and preconditioning, thr…

We use standard physics techniques to model trading and price formation in a market under the assumption that order arrival and cancellations are Poisson random processes. This model makes testable predictions for the most basic properties of a market, such as the diffusion rate of prices, which is the standard measure…

Many research fields codify their findings in standard formats, often by reporting correlations between quantities of interest. But the space of all testable correlates is far larger than scientific resources can currently address, so the ability to accurately predict correlations would be useful to plan research and a…

New method discovers context effects in choice data.

Learning shrinks hard tail, improving inference performance.

New method estimates causal effects without knowing graph structure.

Criteria for extending degree-2 Azumaya algebras with C2-actions over curves.

Paper formalizes continual semi-supervised anomaly detection, showing promising results.

Typical cohorts in brain imaging studies are not large enough for systematic testing of all the information contained in the images. To build testable working hypotheses, investigators thus rely on analysis of previous work, sometimes formalized in a so-called meta-analysis. In brain imaging, this approach underlies th…

ZDP detects drift in large language models without labels, proving key theorems and metrics.

Surprise-based learning allows agents to rapidly adapt to non-stationary stochastic environments characterized by sudden changes. We show that exact Bayesian inference in a hierarchical model gives rise to a surprise-modulated trade-off between forgetting old observations and integrating them with the new ones. The mod…

New framework limits testing algorithmic stability under computational constraints.

We provide a microfoundation for linear price impact models in a stationary market.

New algorithm reduces contamination in supervised learning.

This paper studies structure detection problems in high temperature ferromagnetic (positive interaction only) Ising models. The goal is to distinguish whether the underlying graph is empty, i.e., the model consists of independent Rademacher variables, versus the alternative that the underlying graph contains a subgraph…

Paper presents ML-VAMP for efficient multi-layer inference with exact performance analysis.

Tests whether a treatment's effect is fully mediated by observed outcomes and identifies causal mechanisms.

Detect hidden confounding in observational data using multiple environments.

NFT royalties boost creator earnings by sharing risk, reducing info asymmetry, and enabling price discrimination.

The two-dimensional renormalization group acting as the Ricci flow $Λ\frac{\partial}{\partialΛ} g_{μν} = R_{μν}$ produces a specific 1+3 dimensional space-time metric which describes an expanding universe that starts with a big bang $a \sim t^{\scriptscriptstyle 1/\sqrt3}$ then decelerates until $z=0.2$ then accelerate…