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.
RayS attack improves hard-label adversarial attacks by reducing query complexity and identifying false robust models.
New PL method improves ASR accuracy without pseudo-labels.
Efficient method for generating adversarial examples with limited query budget.
This work connects hardness of approximation and learning.
We study the problem of attacking a machine learning model in the hard-label black-box setting, where no model information is revealed except that the attacker can make queries to probe the corresponding hard-label decisions. This is a very challenging problem since the direct extension of state-of-the-art white-box at…
Although deep convolutional neural networks achieve state-of-the-art performance across nearly all image classification tasks, their decisions are difficult to interpret. One approach that offers some level of interpretability by design is \textit{hard attention}, which uses only relevant portions of the image. However…
The major challenge of learning from multi-label data has arisen from the overwhelming size of label space which makes this problem NP-hard. This problem can be alleviated by gradually involving easy to hard tags into the learning process. Besides, the utilization of a diversity maintenance approach avoids overfitting …
Rotation invariant algorithms fail with hard labels sampled from sparse targets.
The goal of semi-supervised learning is to improve supervised classifiers by using additional unlabeled training examples. In this work we study a simple self-learning approach to semi-supervised learning applied to the least squares classifier. We show that a soft-label and a hard-label variant of self-learning can be…
Retraining with predicted labels improves model accuracy in noisy settings.
Distillation works even with hard labels from overparameterized teacher, leading to better performance.
In Domain Adaptation (DA), the category-relevant losses usually occupy a dominant position, while they are usually built with hard or soft labels in existing models. We observed that hard labels are overconfident due to hard samples existed, and soft labels are ambiguous as too many small noisy probabilities involved, …
We study the most practical problem setup for evaluating adversarial robustness of a machine learning system with limited access: the hard-label black-box attack setting for generating adversarial examples, where limited model queries are allowed and only the decision is provided to a queried data input. Several algori…
Efficient algorithms learn from coarse labels instead of fine grained ones.
We propose a novel approach for estimating the difficulty and transferability of supervised classification tasks. Unlike previous work, our approach is solution agnostic and does not require or assume trained models. Instead, we estimate these values using an information theoretic approach: treating training labels as …
Training accurate deep neural networks (DNNs) in the presence of noisy labels is an important and challenging task. Though a number of approaches have been proposed for learning with noisy labels, many open issues remain. In this paper, we show that DNN learning with Cross Entropy (CE) exhibits overfitting to noisy lab…
Paper solves NP-hard sparse mixed linear regression problem with provable guarantees.
Learning using privileged information is an attractive problem setting that helps many learning scenarios in the real world. A state-of-the-art method of Gaussian process classification (GPC) with privileged information is GPC+, which incorporates privileged information into a noise term of the likelihood. A drawback o…
Simple k-NN filtering improves model accuracy on noisy labels.
Study shows SQ hardness for multiclass linear classification with random noise.
Paper tackles representation learning with limited labels from crowdsourced data.
Paper estimates optimal classification error with soft labels and calibration.
This paper analyzes the difficulty of unsupervised domain adaptation using information theory.
Meta-learning strategy improves few-shot classification performance.
While on some natural distributions, neural-networks are trained efficiently using gradient-based algorithms, it is known that learning them is computationally hard in the worst-case. To separate hard from easy to learn distributions, we observe the property of local correlation: correlation between local patterns of t…
We study the problem of finding a universal (image-agnostic) perturbation to fool machine learning (ML) classifiers (e.g., neural nets, decision tress) in the hard-label black-box setting. Recent work in adversarial ML in the white-box setting (model parameters are known) has shown that many state-of-the-art image clas…
PDO optimizes LLM prompts without labels, improving performance.
Proposes a robust VIB approach using soft labels and mutual info estimation.
We study how to adapt to smoothly-varying ('easy') environments in well-known online learning problems where acquiring information is expensive. For the problem of label efficient prediction, which is a budgeted version of prediction with expert advice, we present an online algorithm whose regret depends optimally on t…
The paper analyzes knowledge distillation in wide neural networks, providing theoretical insights and practical implications.
Proposes a method to generate high-quality candlestick data for financial trading.
SGD-trained neural networks generalize well even with adversarial label noise.
New machine learning model faster, more accurate, and can identify hard-to-classify samples.
We study the problem of recovering the latent ground truth labeling of a structured instance with categorical random variables in the presence of noisy observations. We present a new approximate algorithm for graphs with categorical variables that achieves low Hamming error in the presence of noisy vertex and edge obse…
Proposes a new method for deep ensembles that improves accuracy and calibration.
The generalization and learning speed of a multi-class neural network can often be significantly improved by using soft targets that are a weighted average of the hard targets and the uniform distribution over labels. Smoothing the labels in this way prevents the network from becoming over-confident and label smoothing…
The state-of-the-art solutions for Aspect-Level Sentiment Analysis (ALSA) were built on a variety of deep neural networks (DNN), whose efficacy depends on large amounts of accurately labeled training data. Unfortunately, high-quality labeled training data usually require expensive manual work, and may thus not be readi…
Fine-grained pretraining improves neural network's ability to learn rare features.
Scarcity of labeled data is a bottleneck for supervised learning models. A paradigm that has evolved for dealing with this problem is data programming. An existing data programming paradigm allows human supervision to be provided as a set of discrete labeling functions (LF) that output possibly noisy labels to input in…
Reduces learning periodic neural networks to lattice problems, proving hardness under cryptographic assumptions.
Proposes a weighted conformal approach for cluster label uncertainty.
Two strategies extend multi-label chaining for imprecise probability estimates.
Paper tackles instance-dependent label noise by approximating it with part-dependent noise.
Adaptive sampling improves deep learning efficiency by focusing on informative samples.
The problem of learning from label proportions (LLP) involves training classifiers with weak labels on bags of instances, rather than strong labels on individual instances. The weak labels only contain the label proportion of each bag. The LLP problem is important for many practical applications that only allow label p…
New Gibbs sampling reduces GLMB filtering complexity to linear time.
Simplifies multi-label classification with stochastic sketch strategy.
We consider the problem of computing the best-fitting ReLU with respect to square-loss on a training set when the examples have been drawn according to a spherical Gaussian distribution (the labels can be arbitrary). Let be the population loss of the best-fitting ReLU. We prove: 1. Finding a ReLU wit…