SUMER updates deployed models with new data, improving performance.
problem Updating deployed machine learning models with new data is difficult.
method SUMER uses semi-supervised learning and noise remediation to iteratively retrain models.
result SUMER improves model performance, especially with limited initial training data.
This paper tackles label noise in deep learning for medical image analysis.
problem Label noise impacts deep learning models in medical image analysis.
method Review of state-of-the-art techniques and experiments with label noise in medical datasets.
result Developed new methods to combat label noise in deep models for medical image analysis.
STAG injects noise into graph neural networks to improve performance.
problem Graph neural networks suffer from over-smoothing and limited discrimination.
method Introduces a stochastic aggregation framework (STAG) with adaptive noise injection.
result STAG models correct both over-smoothing and discrimination issues.
Transformer model removes noise from light curves efficiently.
problem Challenges in processing astrophysical light curves due to noise.
method Denoising Time Series Transformer (DTST) model trained with masked objective.
result DTST model excels at removing noise and outliers in time series datasets.
DGPs with variational inference suffer from SNR issues that degrade gradient estimates, leading to unreliable training.
problem SNR issues in gradient estimates for DGPs with variational inference.
method Adapted doubly reparameterized gradient estimators for DGP training.
result Fix improves predictive performance of DGP models.
We analyze anomaly detection class imbalance using a solvable model.
problem Class imbalance hampers anomaly detection performance.
method We use an exact solution of the teacher-student perceptron model through replica theory.
result Optimal train imbalance is often different from 50%, influenced by intrinsic imbalance and data abundance.
In high dimension, it is customary to consider Lasso-type estimators to enforce sparsity. For standard Lasso theory to hold, the regularization parameter should be proportional to the noise level, yet the latter is generally unknown in practice. A possible remedy is to consider estimators, such as the Concomitant/Scale…
Generative adversarial networks (GANs) are a framework that learns a generative distribution through adversarial training. Recently, their class-conditional extensions (e.g., conditional GAN (cGAN) and auxiliary classifier GAN (AC-GAN)) have attracted much attention owing to their ability to learn the disentangled repr…
Node2vec embeddings are unstable and unstable with parameter choices.
problem Stability and robustness of node2vec embeddings for graph classification.
method Analysis of node2vec embeddings from multiple perspectives.
result Node2vec embeddings are unstable with respect to parameter choices.
Diffusion-GAN uses diffusion to improve GAN training stability and realism.
problem Stability and realism issues in training GANs.
method Diffusion-GAN employs a forward diffusion chain to generate Gaussian-mixture distributed instance noise, with adaptive diffusion process and timestep-dependent discriminator.
result Diffusion-GAN produces more realistic images with higher stability and data efficiency.
Improved predictive posterior density estimation through optimized importance sampling.
problem Low signal-to-noise ratio in posterior predictive densities.
method Optimized importance sampling using a test-time variational proxy.
result Significantly improved estimates of predictive posterior densities.
The paper proposes using network science to improve portfolio optimization by reducing noise in covariance estimation.
problem Noise in covariance estimation leads to suboptimal portfolio performance.
method The paper introduces SR-IFN, a network-based method to filter out noise from empirical covariance, enhancing portfolio optimization.
result The SR-IFN network improves portfolio performance by selecting peripheral, diversified assets and inversely weighting them based on centrality.
Develops a machine learning framework for computing most probable paths in stochastic systems.
problem Computing the most probable paths in stochastic dynamical systems.
method Reformulates the boundary value problem of Hamiltonian systems and uses a neural network to solve the Euler-Lagrange equation for the Onsager-Machlup action functional.
result Demonstrates the efficacy and accuracy of the machine learning approach in computing most probable paths for stochastic systems with various types of noise.
Novel algorithm for private analysis of fMRI data.
problem Privacy concerns in collaborative neuroimaging data analysis.
method Differentially private decentralized ICA algorithm using correlated noise.
result Algorithm outperforms existing methods on synthetic and real neuroimaging datasets.
Machine learning, especially deep neural networks, has been rapidly developed in fields including computer vision, speech recognition and reinforcement learning. Although Mini-batch SGD is one of the most popular stochastic optimization methods in training deep networks, it shows a slow convergence rate due to the larg…
Study on how SEU affects neural networks and proposes remedies.
problem Impact of SEU on neural networks' robustness.
method Defined fault models of SEU, defined sensitivity to SIPP, analytically explored network weaknesses, proposed remedies.
result Proposed remedies can mitigate accuracy degradation from 28% to 0.27%.
New method targets deep learning optima with heavy-tailed noise.
problem Heavy-tailed gradient noise biases SGDm optima.
method Fractional Underdamped Langevin Dynamics (FULD) for heavy-tailed noise.
result FULD targets Gibbs distribution matching optima of cost function.
REMEDI improves neural entropy estimation across various tasks.
problem Challenges in estimating information theoretic quantities in high-dimensional data.
method Combines minimization of cross-entropy with estimation of deviation from data density.
result Improves accuracy in entropy estimation on synthetic and natural data.
A brief historical perspective is first given concerning financial crashes, - from the 17th till the 20th century. In modern times, it seems that log periodic oscillations are found before crashes in several financial indices. The same is found in sand pile avalanches on Sierpinski gaskets. A discussion pertains to the…
A new method for sampling high-dimensional distributions overcomes overfitting.
problem Overfitting in energy-based models during gradient descent.
method Mean-field microcanonical gradient descent, which samples multiple data points simultaneously.
result The method reduces entropy loss while maintaining likelihood fit, improving overfitting issues.
Generative model initializes 2-layer network weights for small datasets.
problem Approximating functions with 2-layer networks using small datasets and gradient-based training.
method Initialize hidden weights with a learned proposal distribution parameterized as a deep generative model. Refine with gradient-based post-processing and regularization.
result Demonstrates effectiveness of the approach with numerical examples.
In high dimensional settings, sparse structures are crucial for efficiency, both in term of memory, computation and performance. It is customary to consider ℓ1 penalty to enforce sparsity in such scenarios. Sparsity enforcing methods, the Lasso being a canonical example, are popular candidates to address high dim…
New deep learning models inspired by fuzzy logic are more robust to adversarial attacks.
problem Flawed generalization in deep neural networks leading to adversarial examples.
method Inspired by fuzzy logic, new architectures combining alternative design elements.
result New models are more robust to adversarial examples and noise.
Lasso is a seminal contribution to high-dimensional statistics, but it hinges on a tuning parameter that is difficult to calibrate in practice. A partial remedy for this problem is Square-Root Lasso, because it inherently calibrates to the noise variance. However, Square-Root Lasso still requires the calibration of a t…
Proposes R2LDA for improved LDA classifier performance.
problem Poor performance of LDA classifiers in small to comparable data sizes.
method Doubly regularized LDA with automatic parameter selection.
result Consistent and effective performance in noisy test data.
Paper presents a probabilistic diagnostic model for identifying and treating supervised learning degradation issues.
problem Degradation problems in supervised learning, including class imbalance, overlapping, small-disjuncts, noisy labels, and sparseness.
method Develops a novel probabilistic diagnostic model to identify and treat degradation issues in supervised learning.
result Early and correct diagnosis of degradation issues allows for selecting appropriate remediation treatments and unbiased performance metrics.
Interval Neural Networks detect instabilities in image reconstructions.
problem Detecting instabilities in deep learning image reconstructions.
method Employed uncertainty quantification methods with Interval Neural Networks.
result Interval Neural Networks effectively reveal image reconstruction instabilities.
Nested Cavity Classifier (NCC) is a classification rule that pursues partitioning the feature space, in parallel coordinates, into convex hulls to build decision regions. It is claimed in some literatures that this geometric-based classifier is superior to many others, particularly in higher dimensions. First, we give …
We use automatic speech recognition to assess spoken English learner pronunciation based on the authentic intelligibility of the learners' spoken responses determined from support vector machine (SVM) classifier or deep learning neural network model predictions of transcription correctness. Using numeric features produ…
Particle-optimization-based sampling (POS) is a recently developed effective sampling technique that interactively updates a set of particles. A representative algorithm is the Stein variational gradient descent (SVGD). We prove, under certain conditions, SVGD experiences a theoretical pitfall, {\it i.e.}, particles te…
Study on W2S generalization with spurious correlations, proposing remedies.
problem Understanding and improving W2S generalization with spurious correlations.
method Theoretical analysis and algorithmic remedies for W2S fine-tuning.
result W2S always happens with sufficient pseudolabels when group fractions match, but may fail otherwise.
This paper has been withdrawn by the authors, because it has been made obsolete by the detailed expositions in our papers in arXiv:0812.4885 (the mathematics part) and arXiv:0812.4737 (the economics part).
DUA-D2C refines D2C for better deep learning generalization.
problem Overfitting in deep learning models due to data outliers and noise.
method Dynamic uncertainty-aware aggregation of D2C subsets.
result Significant improvement in generalization performance.
A framework for private prediction sets using conformal prediction and differential privacy.
problem Jointly addressing reliability and privacy in machine learning predictions.
method Split conformal prediction with privatized quantile subroutine.
result Private prediction sets can be generated from privately-trained models.
A new method reduces sample complexity for meta-learning.
problem Efficiently learn new tasks with minimal data.
method Alternating minimization method (MLLAM) for linear regression tasks.
result MLLAM achieves nearly-optimal estimation error with Ω(logd) samples per task. We develop new techniques in the theory of convex surfaces to prove complete classification results for tight contact structures on lens spaces, solid tori, and T^2 X I. Erratum: In this note we seek to remedy errors which appeared in version 2 and were propagated in subsequent papers.
Sigmoid-type networks avoid vanishing gradients with regularization and rescaling.
problem Vanishing gradients in sigmoid-type networks.
method Mathematical arguments and two remedies: regularization and rescaling.
result Demonstrates effectiveness of regularization and rescaling in practice.
A simple strategy prevents negative transfer in transfer learning.
problem Negative transfer in transfer learning where source representations harm target performance.
method Residual feature integration with a trainable target-side encoder.
result The method provably prevents negative transfer with theoretical guarantees.
The problem of inhomogeneous cluster densities has been a long-standing issue for distance-based and density-based algorithms in clustering and anomaly detection. These algorithms implicitly assume that all clusters have approximately the same density. As a result, they often exhibit a bias towards dense clusters in th…
Every smooth fiber bundle admits a complete (Ehresmann) connection. This result appears in several references, with a proof on which we have found a gap, that does not seem possible to remedy. In this note we provide a definite proof for this fact, explain the problem with the previous one, and illustrate with examples…
The vector of periodic, compound returns of a typical investment portfolio is almost never a convex combination of the return vectors of the securities in the portfolio. As a result the ex post version of Harry Markowitz's "standard mean-variance portfolio selection model" does not apply to compound return data. We pro…
Transformers without skip connections collapse token representations to a single direction.
problem Rapid convergence of token representations to a single direction in self-attention-only Transformers.
method Analysis of layer normalization, residual connections, and multi-head attention mechanisms.
result Residual connections prevent rank collapse in real Transformers, while MLPs generate new feature directions.
Bayesian deep neural networks converge to processes with α-stable marginals under infinite variance weights.
problem Representation learning in deep kernel processes is hindered by deterministic covariance kernels.
method Showed convergence to α-stable processes with conditionally Gaussian representations in infinite-width networks.
result Conditional random covariance kernels can be recursively linked, even if the process is α-stable.
Paper proposes a new model for noisy tensor completion.
problem Handling noise in tensor completion.
method Tensor ring nuclear norm (TRNN) and least-squares estimator.
result Effective recovery of noisy incomplete tensor data.
Unified AI system for data quality control and governance in regulated environments.
problem Isolated data quality control steps in existing systems.
method AI-driven framework integrating rule-based, statistical, and AI methods.
result Empirical gains in anomaly detection, reduced manual remediation, improved auditability.
Despite their impressive performance in many tasks, deep neural networks often struggle at relational reasoning. This has recently been remedied with the introduction of a plug-in relational module that considers relations between pairs of objects. Unfortunately, this is combinatorially expensive. In this extended abst…
A novel ensemble classifier improves vibration-based quality monitoring accuracy.
problem Developing high accuracy classification methods for general datasets.
method Dempster-Shafer theory of evidence with three remedies for conflicting evidences.
result The proposed ensemble classifier outperforms state-of-the-art fusion techniques.
Develops a theoretical framework for scalable Gaussian Process regression methods.
problem Limited scalability of Gaussian Process regression for large datasets.
method Introduces and analyzes Nearest Neighbour Gaussian Process (NNGP) and scalable GPnn methods.
result Derives almost sure pointwise limits for predictive criteria and proves risk minimax rates.