A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
Clinical AI models fail to transfer between sites due to site-specific practices.
A site-specific Gordian distance between two spatial embeddings of an abstract graph is the minimal number of crossing changes from one to another where each crossing change is performed between two previously specified abstract edges of the graph. It is infinite in some cases. We determine the site-specific Gordian di…
Site-specific recombination is an enzymatic process where two sites of precise sequence and orientation along a circle come together, are cleaved, and the ends are recombined. Site-specific recombination on a knotted substrate produces another knot or a two-component link depending on the relative orientation of the si…
Site-specific recombination on supercoiled circular DNA molecules can yield a variety of knots and catenanes. Twist knots are some of the most common conformations of these products and they can act as substrates for further rounds of site-specific recombination. They are also one of the simplest families of knots and …
We develop a topological model of site-specific recombination that applies to substrates which are the connected sum of two torus links of the form . Then we use our model to prove that all knots and links that can be produced by site-specific recombination on such substrates are contained in one of two…
Method improves robustness and generalizability of CATE estimation.
Knowledge-based planning (KBP) is an automated approach to radiation therapy treatment planning that involves predicting desirable treatment plans before they are then corrected to deliverable ones. We propose a generative adversarial network (GAN) approach for predicting desirable 3D dose distributions that eschews th…
Bio-oil molecule assessment is essential for the sustainable development of chemicals and transportation fuels. These oxygenated molecules have adequate carbon, hydrogen, and oxygen atoms that can be used for developing new value-added molecules (chemicals or transportation fuels). One motivation for our study stems fr…
We develop a model characterizing all possible knots and links arising from recombination starting with a twist knot substrate, extending previous work of Buck and Flapan. We show that all knot or link products fall into three well-understood families of knots and links, and prove that given a positive integer , the…
Paper proposes a privacy-preserving method for estimating complex models.
We develop a topological model of knots and links arising from a single (or multiple processive) round(s) of recombination starting with an unknot, unlink, or (2,m)-torus knot or link substrate. We show that all knotted or linked products fall into a single family, and prove that the size of this family grows linearly …
Understanding and accurately predicting within-field spatial variability of crop yield play a key role in site-specific management of crop inputs such as irrigation water and fertilizer for optimized crop production. However, such a task is challenged by the complex interaction between crop growth and environmental and…
Generative model learns wireless channel distributions efficiently.
Meta-analysis improves personalized treatment rules across multiple sites.
Flooding is a destructive and dangerous hazard and climate change appears to be increasing the frequency of catastrophic flooding events around the world. Physics-based flood models are costly to calibrate and are rarely generalizable across different river basins, as model outputs are sensitive to site-specific parame…
New method combines regional HIV prevention trial data without sharing individual patient info.
fedCI and fedCI-IOD enable federated causal discovery across diverse datasets with privacy and power enhancements.
New ZIPLN model accounts for zero-inflation in multivariate count data.
Access to sufficient annotated data is a common challenge in training deep neural networks on medical images. As annotating data is expensive and time-consuming, it is difficult for an individual medical center to reach large enough sample sizes to build their own, personalized models. As an alternative, data from all …
The goal of this paper is to assess the utility of Reduced-Order Models (ROMs) developed from 3D physics-based models for predicting transient thermal power output for an enhanced geothermal reservoir while explicitly accounting for uncertainties in the subsurface system and site-specific details. Numerical simulations…
GOPSA optimizes EEG data for cross-site age prediction, improving performance on multiple metrics.
Paper establishes no-regret property for practical EGO optimization.
Gluon optimizes LMO-based methods for large-scale tasks, improving performance and theory-practice gap.
Semi-supervised wrapper methods are concerned with building effective supervised classifiers from partially labeled data. Though previous works have succeeded in some fields, it is still difficult to apply semi-supervised wrapper methods to practice because the assumptions those methods rely on tend to be unrealistic i…
Overlay framework simplifies exotic derivative pricing.
The study compares machine learning models for depression detection and highlights the importance of feature selection.
Study assesses environmental management accounting practices in Bangladesh.
New framework provides privacy guarantees for practical federated learning.
A practical guide to Variational Bayes methods.
In the Best- identification problem (Best--Arm), we are given stochastic bandit arms with unknown reward distributions. Our goal is to identify the arms with the largest means with high confidence, by drawing samples from the arms adaptively. This problem is motivated by various practical applications and…
NTK theory fails to predict practical behavior of large-width neural networks.
In several realistic situations, an interactive learning agent can practice and refine its strategy before going on to be evaluated. For instance, consider a student preparing for a series of tests. She would typically take a few practice tests to know which areas she needs to improve upon. Based of the scores she obta…
BLAE solves batched linear bandits with optimal regret and practical performance.
A practical one-shot federated learning algorithm for cross-silo setting.
Privacy policies are legal documents that describe how a website will collect, use, and distribute a user's data. Unfortunately, such documents are often overly complicated and filled with legal jargon; making it difficult for users to fully grasp what exactly is being collected and why. Our solution to this problem is…
Study highlights how lazy data practices in fair ML research can unfairly impact minority groups.
This thesis improves practical reinforcement learning methods with robustness, scalability, and efficiency.
Good sparse approximations are essential for practical inference in Gaussian Processes as the computational cost of exact methods is prohibitive for large datasets. The Fully Independent Training Conditional (FITC) and the Variational Free Energy (VFE) approximations are two recent popular methods. Despite superficial …
Proposes guidelines for developing medical AI products.
New algorithm ensures global convergence in deep neural networks beyond NTK regime.
Practical algorithm for contextual bandits with large action spaces.
RQMC improves QMC by providing practical error bounds for financial applications.
New method AnInfoNCE uncovers latent factors in contrastive learning with practical variability.
PRoA assesses deep learning robustness against practical functional perturbations.
Addresses theoretical and practical aspects of Gaussian differential privacy.
Differentially private learning on real-world data poses challenges for standard machine learning practice: privacy guarantees are difficult to interpret, hyperparameter tuning on private data reduces the privacy budget, and ad-hoc privacy attacks are often required to test model privacy. We introduce three tools to ma…
Stochastic Gradient Descent (SGD) is widely used in machine learning problems to efficiently perform empirical risk minimization, yet, in practice, SGD is known to stall before reaching the actual minimizer of the empirical risk. SGD stalling has often been attributed to its sensitivity to the conditioning of the probl…
Despite widespread interest and practical use, the theoretical properties of random forests are still not well understood. In this paper we contribute to this understanding in two ways. We present a new theoretically tractable variant of random regression forests and prove that our algorithm is consistent. We also prov…