A new method quantizes output space for multi-target regression.
problem Predicting multiple continuous targets using shared predictors.
method MRQ method that quantizes output space to model dependencies and scale.
result MRQ achieves high scalability and competitive accuracy.
Paper presents MTTDSC for better target-specific sentiment classification.
problem Improving accuracy in detecting and aggregating sentiments towards specific targets in social media.
method MTTDSC uses a multi-task learning approach with an auxiliary task for passage-level sentiment classification and a main task for target-specific sentiment classification.
result MTTDSC outperforms state-of-the-art baselines in sentiment classification.
Proposes a new online learning strategy for multi-target regression in data streams.
problem Challenges in learning from high-throughput data streams, especially in multi-target regression.
method Extends existing online decision tree learning algorithm to consider inter-target dependencies.
result SST-HT presents superior predictive accuracy compared to state-of-the-art algorithms.
New algorithm predicts multiple types of outputs with dependencies.
problem Predicting multiple diverse types of outputs.
method Problem transformation method combined with component-wise boosting.
result Sparse and interpretable learning of dependencies between targets.
Gradient descent reshapes the function space of neural networks.
problem Understanding how feature learning affects the function space of neural networks.
method Characterized the evolution of the feature space during training using a two-layer neural network.
result Gradient descent induces a data-adaptive deformation that selectively enhances signal-aligned directions.
This paper quantifies how hard it is to identify specific data points in machine learning models.
problem Quantifying the difficulty of identifying specific data points in machine learning models.
method Characterizing optimal attacks and privacy defences, deriving impacts of noise and misspecification, and proposing a new covariance attack.
result The Mahalanobis distance explains the hardness of fixed-target membership inference attacks.
This paper tackles target-dependent label complexity gap in active learning.
problem Target-dependent label complexity gap in Agnostic Active Learning.
method Introduces a novel distribution-splitting strategy based on number density to reduce label complexity and error rate.
result Provides theoretical guarantees and practical advantages for reducing label complexity and error rate.
We investigate structured sparsity methods for variable selection in regression problems where the target depends nonlinearly on the inputs. We focus on general nonlinear functions not limiting a priori the function space to additive models. We propose two new regularizers based on partial derivatives as nonlinear equi…
Transfer learning for causal forest
problem Estimating CATE in a causal forest
method Offset method adapted to causal context
result Bound on CATE error
Study SGD dynamics in sequence models, revealing training phases and influence of sequence length.
problem Understanding SGD in sequence models like attention networks.
method Derived closed-form population loss and analyzed SGD dynamics for SSI models.
result Two distinct training phases: escape from uninformative initialization and alignment with target subspace.
We consider support recovery in the quadratic logistic regression setting - where the target depends on both p linear terms xi and up to p2 quadratic terms xixj. Quadratic terms enable prediction/modeling of higher-order effects between features and the target, but when incorporated naively may involve solvi…
FIESTA optimizes model selection by efficiently evaluating multiple splits and seeds.
problem Inefficient model selection leading to unreliable performance comparisons.
method Adaptive bandit algorithms to determine optimal number of data splits and random seeds.
result Significantly reduces model evaluations while ensuring correct optimal model selection.
In this paper we investigate novel applications of a new class of equations which we call time-delayed backward stochastic differential equations. Time-delayed BSDEs may arise in finance when we want to find an investment strategy and an investment portfolio which should replicate a liability or meet a target depending…
CSD improves goodness-of-fit testing for higher-order dependence.
problem Insensitivity of standard KSDs to higher-order dependence features like tail dependence.
method Introduces Copula-Stein Discrepancy (CSD) that targets dependence geometry directly on copula density.
result CSD is sensitive to differences in tail dependence coefficients and metrizes weak convergence of copula distributions.
Study compares methods for recovering latent risk-neutral densities from option prices, finding DeepONet effective.
problem Accurately recovering latent risk-neutral densities from option prices is challenging.
method Two benchmarks and various methods (lognormal mixture, DeepONet, quote transformer) are used to compare recovery accuracy.
result DeepONet outperforms other methods in reducing error on latent density recovery.
Introduces Fitzpatrick losses, tighter than Fenchel-Young losses.
problem Improving loss functions for machine learning.
method Introduces Fitzpatrick losses based on the Fitzpatrick function.
result Fitzpatrick losses are tighter than Fenchel-Young losses.
This paper studies universal rates of ERM for binary classification under agnostic learning.
problem The challenge of achieving universal rates of ERM for binary classification under agnostic learning.
method The paper explores the agnostic universal rates of ERM for binary classification, revealing three possible rates: e−n, o(n−1/2), or arbitrarily slow. result The paper provides a complete characterization of which concept classes fall into each of the three categories of agnostic universal rates.
Bayesian methods detect significant IIA violations in similarity choice data.
problem Detecting IIA violations in similarity choice data complicates classical models.
method Proposed two statistical methods: classical goodness-of-fit test and Bayesian PPC.
result Significant IIA violations confirmed in both datasets, driven by context effects.
GIRNet tackles multi-tasking with mixed domain sequences, improving sentiment and tagging tasks.
problem Labeling sequences with mixed domain data and position-specific inference.
method Unified position-sensitive multi-task RNN architecture with gated state sequences from auxiliary data.
result GIRNet achieves new state-of-the-art performance in sentiment classification, POS tagging, and target position-sensitive annotation.
CogMol designs novel drug-like molecules for SARS-CoV-2 targets.
problem Designing efficient drugs for novel viral proteins.
method End-to-end framework combining VAE, controlled sampling, and predictors.
result Highly selective and affinity molecules for SARS-CoV-2 targets.
Breaks circular dependency in synthetic option pricing with a novel model.
problem Circular dependency in implied volatility limits synthetic data for machine learning and risk analysis.
method Uses a Jump-Hidden Markov Model to generate price paths and a modified Heston process to convert paths into implied volatility.
result Framework generates realistic synthetic American option prices without external calibration.
Wide neural networks learn features under μP, identifying weights and decomposing support.
problem Feature learning in wide neural networks under μP. method Proving mean-field limit, characterizing identifiability, sparse-dictionary decomposition, and feature-learning-error decomposition.
result The triple (w∗,Dorb∗,S∗) identifies the natural learning cell of the architecture-data pair (σ,ρ).