Holdout set improves risk score accuracy without biasing predictions.
problem Updating risk scores can lead to biased estimates when directly applied.
method Use a holdout set of non-intervention population to update risk scores.
result Optimal holdout size reduces adverse outcomes to optimal level.
Updating predictive models introduces bias, leading to miscalibration.
problem Bias introduced when updating predictive models after interventions.
method Proposed a causal framework and partially observed Markov decision process.
result Successive predictive scores may converge to undesirable outcomes.
BaM improves BBVI by optimizing a score-based divergence, leading to faster convergence.
problem Slow convergence of black-box variational inference methods.
method Batch and match (BaM) approach based on a score-based divergence.
result BaM converges exponentially quickly to the target mean and covariance.
A new filter reduces density fitting to a linear solve, improving performance on nonlinear systems.
problem Nonlinear Bayesian filtering challenges in representing belief distributions.
method Combines score matching with Stein's identity to avoid partition function evaluation.
result The Score Kalman Filter (SKF) outperforms existing methods on nonlinear systems.
Improves BBVI for high-dimensional Gaussian approximations by using low-rank approximations.
problem Scalability issues with BBVI for high-dimensional multivariate Gaussian approximations.
method Extends BaM framework to handle full covariance matrices by integrating patch step for low-rank parameterization.
result Shows improved efficiency and scalability on synthetic and real-world high-dimensional inference problems.
Iterative tilting fine-tunes diffusion models for reward-tilted distributions.
problem Fine-tuning diffusion models for reward-tilted distributions.
method Decomposes large reward tilts into smaller, tractable tilts via first-order Taylor expansion, avoiding backpropagation.
result Validated on a two-dimensional Gaussian mixture, achieving exact closed-form solutions.
Test-time training adapts a pretrained model to each prompt via parameter updates, improving accuracy under pretraining-to-test distribution shifts.
problem Improving accuracy of pretrained models under distribution shifts.
method Explaining TTT behavior through a decision-theoretic lens.
result TTT reduces prediction error when updates are spectrally matched to the prompt's signal-to-noise ratio and aligned with query-relevant eigen-directions.
Novel method for nonlinear data assimilation using Langevin sampling.
problem Nonlinear data assimilation challenges in Bayesian filtering.
method Score-based sequential Langevin sampling (SSLS) with dynamic models and annealing.
result Asymptotic stability and error bounds for local posterior sampling.
We introduce a simple (one line of code) modification to the Generative Adversarial Network (GAN) training algorithm that materially improves results with no increase in computational cost: When updating the generator parameters, we simply zero out the gradient contributions from the elements of the batch that the crit…
To continuously improve quality and reflect changes in data, machine learning applications have to regularly retrain and update their core models. We show that a differential analysis of language model snapshots before and after an update can reveal a surprising amount of detailed information about changes in the train…
Designing deterministic denominators for SGLD stabilizes large drifts.
problem Stabilizing large drifts in SGLD
method Using state-dependent envelopes and empirical quantiles for activation thresholds
result Proxy-quantile denominators are close to oracle-score behavior and improve deterministic taming choices
T-PSDA improves speaker recognition accuracy on toroidal submanifolds.
problem Improving speaker recognition accuracy on hypersphere embeddings.
method Extends PSDA to model within and between-speaker variabilities in toroidal submanifolds of the hypersphere.
result T-PSDA achieves accuracy on par with cosine scoring on VoxCeleb and large accuracy gains on NIST SRE'21.
FLOP algorithm speeds up causal structure learning for linear models.
problem Efficiently learning causal structures from discrete data.
method FLOP algorithm combines fast parent selection and iterative score updates.
result FLOP finds highly accurate causal structures with near-perfect recovery.
Generalized cross validation (GCV) is one of the most important approaches used to estimate parameters in the context of inverse problems and regularization techniques. A notable example is the determination of the smoothness parameter in splines. When the data are generated by a state space model, like in the spline c…
Model analyzes cooccurrence data for recommender systems and item relevance.
problem High-dimensional cooccurrence data from online platforms.
method Shared parameter Alternating Tweedie (SA-Tweedie) model with Fisher scoring and learning rate adjustment.
result SA-Tweedie model outperforms other methods in optimizing parameters.
Active sampling algorithm improves accuracy of inferred scores from pairwise comparisons.
problem Inference of accurate scores from time-consuming pairwise comparisons.
method Approximate message passing and expected information gain maximization.
result ASAP offers the highest accuracy of inferred scores compared to existing methods.
PRISM-FCP improves federated prediction robustness against Byzantine attacks.
problem Byzantine attacks in federated learning.
method Partial model sharing and distance-based maliciousness scores.
result Maintains nominal coverage guarantees under Byzantine attacks.
The advance of modern sensor technologies enables collection of multi-stream longitudinal data where multiple signals from different units are collected in real-time. In this article, we present a non-parametric approach to predict the evolution of multi-stream longitudinal data for an in-service unit through borrowing…
ASVGD accelerates SVGD for efficient sampling.
problem Slow SVGD in high-dimensional sampling.
method Accelerated gradient flow in a metric space of probability densities, using Nesterov's method and momentum-based updates.
result ASVGD outperforms SVGD and other methods in sampling efficiency.
Novel method for SDE calibration from sparse data using neural flows.
problem Calibrating SDEs from sparse, noisy observations.
method Characterization of posterior SDE using neural networks trained to solve a PDE with multiplicative updates.
result Significant improvement in scalability and accuracy compared to classical methods.
A new method for speaker recognition on hyperspheres improves on PLDA's limitations.
problem Improving speaker recognition on hyperspheres with PLDA's limitations.
method Probabilistic Spherical Discriminant Analysis (PSDA) using Von Mises-Fisher distributions.
result PSDA scores are closed-form and can handle various trials, improving over PLDA.
Paper introduces a fast, robust, scalable method for detecting changes in data streams.
problem Detecting changes in data streams efficiently and reliably.
method Bayesian online changepoint detection with provable robustness and scalability.
result The proposed method is more than 10 times faster than previous approaches and provides provable robustness.
Many in-hospital mortality risk prediction scores dichotomize predictive variables to simplify the score calculation. However, hard thresholding in these additive stepwise scores of the form "add x points if variable v is above/below threshold t" may lead to critical failures. In this paper, we seek to develop risk pre…
A new network model combines features of DCBM, LSM, and β-model, using a cancellation trick for parameter estimation.
problem Challenging parameter fitting in network models.
method Introducing a cancellation trick to resolve parameter fitting issues in the logit-DCBM.
result R-SCORE significantly improves community detection over existing methods.
EnSF improves accuracy in tracking high-dimensional nonlinear systems.
problem Low accuracy in high-dimensional, nonlinear filtering problems.
method Score-based diffusion model, mini-batch Monte Carlo estimator.
result EnSF outperforms state-of-the-art methods in tracking high-dimensional systems.
Improved MUSE boosts performance and reduces error in Bayesian inference.
problem Hierarchical Bayesian inference problems
method Implicit differentiation applied to MUSE algorithm
result Significant speedup and improved accuracy compared to Hamiltonian Monte Carlo
Model analyzes RFQ markets using stochastic control to optimize dealer performance and inventory.
problem Optimizing market making in aggregator-routed RFQ markets with varying dealer performance scores.
method Two-tier stochastic control model that separates RFQ-level price competition from macro routing.
result Optimal controls can be expressed through derivatives of reduced Hamiltonians, leading to interpretable mappings from optimal win probabilities to optimal offsets.
Dual training method for EBMs with overparametrized neural networks.
problem Training EBMs with non-convex energies is challenging.
method Derive variational principles and dual GDA algorithm for feature-learning regime.
result Dual GDA algorithm performs best with similar time scales for features and particles.
Optimal sampling reduces power grid data analysis costs.
problem Efficient online analysis of high-speed, correlated IoT data.
method D-optimality criterion-based sampling methods combining Bernoulli and leverage score sampling.
result Leverage score sampling improves computational efficiency and outperforms benchmarks.
In this paper, we propose a method for training neural networks when we have a large set of data with weak labels and a small amount of data with true labels. In our proposed model, we train two neural networks: a target network, the learner and a confidence network, the meta-learner. The target network is optimized to…
DreamPropeller accelerates text-to-3D generation by 4.7x with minimal loss in quality.
problem Long generation times in text-to-3D generation algorithms degrade user experience.
method DreamPropeller uses Picard iterations generalized for non-ODE paths to accelerate parallel sampling.
result Empirically achieves up to 4.7x speedup with negligible quality loss.
Model forgets examples; this research predicts which ones to replay.
problem Language models forget examples during updates, leading to errors.
method Train forecasting models to predict which examples will be forgotten.
result Forecasting models can reduce forgetting of upstream pretraining examples.
Proposes a method to improve learning when training data is not representative.
problem Improving supervised learning when training data is not representative (covariate shift).
method Conditioning on propensity scores to balance covariates within strata.
result Significantly improved target prediction and AUC (0.958) on supernovae classification challenge.
The paper develops a method for self-normalized inference in adaptive experiments.
problem Adaptive experiments require a fixed horizon for ATE estimation, but propensities can change.
method The method uses self-normalized martingale limit theory to estimate ATE.
result The Studentized statistic is asymptotically N(0,1) at the prespecified horizon.
Gradient filters track moving parameters under noisy data and misspecification.
problem Tracking multidimensional time-varying parameters under noisy observations and model misspecification.
method Gradient-based filters update parameters using the gradient of a postulated objective function, evaluated at either the predicted or updated parameters.
result Novel sufficient conditions for exponential stability of the filtered parameter path, and finite-sample and asymptotic mean squared error bounds.
P-OCS detects OOD samples in a low-dimensional subspace, outperforming existing methods.
problem Efficient OOD detection for deep learning models in open-world environments.
method P-OCS operates in the orthogonal complement of the principal subspace, applying a single projected perturbation.
result P-OCS achieves state-of-the-art OOD detection with negligible computational cost and without requiring model retraining.
ScoreStop uses gradient tests to stop gradient boosting early.
problem Overfitting in gradient boosted decision trees.
method ScoreStop uses a functional score test based on gradients to stop boosting.
result ScoreStop is competitive with loss-based early stopping methods.
DDO-RM improves reward-based policies by converting reward scores into a target distribution.
problem Improving reward-based policies when the reward function is simpler than the policy.
method Converts reward scores into a target distribution and uses KL-regularized mirror-descent updates.
result DDO-RM outperforms DPO in pair accuracy and mean margin.
DiffusionBlocks trains neural networks by breaking them into independent blocks, reducing memory usage.
problem Memory bottlenecks in end-to-end neural network training.
method Transforming transformer-based networks into independent trainable blocks via a denoising process.
result Independent block-wise training matches end-to-end training performance while reducing memory requirements.
CPATTA uses conformal prediction for efficient test-time adaptation.
problem Low data selection efficiency in existing ATTA methods.
method Conformal Prediction, online weight-update algorithm, domain-shift detector, staged update scheme.
result CPATTA consistently outperforms state-of-the-art methods by 5% in accuracy.
Signals coming from multivariate higher order conditional moments as well as the information contained in exogenous covariates, can be effectively exploited by rational investors to allocate their wealth among different risky investment opportunities. This paper proposes a new flexible dynamic copula model being able t…
DP-SGD can update fewer coordinates while maintaining privacy.
problem How to update fewer coordinates in DP-SGD without losing optimization signal.
method TP-TopK (Two-Phase TopK DP-SGD), a two-phase method for coordinate-sparse private training.
result Private training can update fewer coordinates without losing optimization signal, scaling noise with active dimension \(k\) instead of full dimension \(d\).
SRMC framework reduces Monte Carlo variance by history-based sampling in high-dimensional spaces.
problem Efficient sampling in high-dimensional discrete or continuous state spaces.
method Score-Repellent Monte Carlo (SRMC) framework that summarizes history through running average of score evaluations.
result Improves estimator variance and mode coverage with constant memory usage.
New analysis shows how cross-entropy training shapes attention in transformers.
problem Understanding how gradient-based learning creates the required internal geometry in transformers.
method Developed a first-order analysis of cross-entropy training effects on attention scores and values in a transformer attention head.
result Introduced an advantage-based routing law and responsibility-weighted update for attention scores and values, respectively.
This study compares neural networks, SVM, and decision trees for corporate credit rating predictions.
problem Predicting corporate credit ratings using machine learning methods.
method Applied four machine learning techniques (Bagged Decision Trees, Random Forest, SVM, MLP) to credit rating datasets.
result Decision tree-based models outperformed other techniques in terms of 'Notch Distance' measure.
A new method optimizes neural sequence models for better task performance.
problem Training neural sequence models with maximum likelihood estimation ignores task losses.
method Maximum likelihood guided parameter search (MGS) in the parameter space.
result MGS optimizes sequence-level losses, reducing repetition and non-termination.
W-Flow generates images in one step, faster and better than multi-step methods.
problem Efficiently generating images from a simple reference distribution to a target data distribution.
method W-Flow uses Wasserstein gradient flows to transform the reference distribution to the target distribution in a single step, trained with Sinkhorn divergence.
result W-Flow achieves state-of-the-art results in ImageNet 256imes256 generation with improved mode coverage and faster sampling. We address the problem of unsupervised disentanglement of discrete and continuous explanatory factors of data. We first show a simple procedure for minimizing the total correlation of the continuous latent variables without having to use a discriminator network or perform importance sampling, via cascading the informat…