Search Results for author:
Found 35 papers, 8 papers with code
Online Policy Learning from Offline Preferences
To address this problem, the present study introduces a framework that consolidates offline preferences and \emph{virtual preferences} for PbRL, which are comparisons between the agent's behaviors and the offline data.
Referee-Meta-Learning for Fast Adaptation of Locational Fairness
When dealing with data from distinct locations, machine learning algorithms tend to demonstrate an implicit preference of some locations over the others, which constitutes biases that sabotage the spatial fairness of the algorithm.
Online Structured Prediction with Fenchel--Young Losses and Improved Surrogate Regret for Online Multiclass Classification with Logistic Loss
We extend the exploit-the-surrogate-gap framework to online structured prediction with \emph{Fenchel--Young losses}, a large family of surrogate losses including the logistic loss for multiclass classification, obtaining finite surrogate regret bounds in various structured prediction problems.
Self-attention Networks Localize When QK-eigenspectrum Concentrates
To this end, we characterize the notion of attention localization by the eigenspectrum of query-key parameter matrices and reveal that a small eigenspectrum variance leads attention to be localized.
Embarrassingly Simple Text Watermarks
LLMs can generate texts that cannot be distinguished from human-written texts.
Necessary and Sufficient Watermark for Large Language Models
Although existing watermarking methods have successfully detected texts generated by LLMs, they significantly degrade the quality of the generated texts.
Feature Normalization Prevents Collapse of Non-contrastive Learning Dynamics
While learned representations may collapse into a single point due to the lack of the repulsive force at first sight, Tian et al. (2021) revealed through the learning dynamics analysis that the representations can avoid collapse if data augmentation is sufficiently stronger than regularization.
Estimating Treatment Effects Under Heterogeneous Interference
One popular application of this estimation lies in the prediction of the impact of a treatment (e. g., a promotion) on an outcome (e. g., sales) of a particular unit (e. g., an item), known as the individual treatment effect (ITE).
Dynamic Model Agnostic Reliability Evaluation of Machine-Learning Methods Integrated in Instrumentation & Control Systems
The method, referred to as the Laplacian distributed decay for reliability (LADDR), determines the difference between the operational and training datasets, which is used to calculate a prediction's relative reliability.
Unbalanced Optimal Transport for Unbalanced Word Alignment
Monolingual word alignment is crucial to model semantic interactions between sentences.
BEVStereo++: Accurate Depth Estimation in Multi-view 3D Object Detection via Dynamic Temporal Stereo
Bounded by the inherent ambiguity of depth perception, contemporary multi-view 3D object detection methods fall into the performance bottleneck.
STORM-GAN: Spatio-Temporal Meta-GAN for Cross-City Estimation of Human Mobility Responses to COVID-19
While deep learning approaches outperform conventional estimation techniques on tasks with abundant training data, the continuously evolving pandemic poses a significant challenge to solving this problem due to data nonstationarity, limited observations, and complex social contexts.
Robust computation of optimal transport by $β$-potential regularization
Optimal transport (OT) has become a widely used tool in the machine learning field to measure the discrepancy between probability distributions.
Will Large-scale Generative Models Corrupt Future Datasets?
These trends lead us to a research question: "\textbf{will such generated images impact the quality of future datasets and the performance of computer vision models positively or negatively?}"
Momentum Tracking: Momentum Acceleration for Decentralized Deep Learning on Heterogeneous Data
In this study, we propose Momentum Tracking, which is a method with momentum whose convergence rate is proven to be independent of data heterogeneity.
BEVStereo: Enhancing Depth Estimation in Multi-view 3D Object Detection with Dynamic Temporal Stereo
To this end, we introduce an effective temporal stereo method to dynamically select the scale of matching candidates, enable to significantly reduce computation overhead.
Ranked #11 on
3D Object Detection
on nuScenes Camera Only
Approximating 1-Wasserstein Distance with Trees
In this paper, we aim to approximate the 1-Wasserstein distance by the tree-Wasserstein distance (TWD), where TWD is a 1-Wasserstein distance with tree-based embedding and can be computed in linear time with respect to the number of nodes on a tree.
Quantitative Evaluation of Common Cause Failures in High Safety-significant Safety-related Digital Instrumentation and Control Systems in Nuclear Power Plants
This paper proposes a Platform for Risk Assessment of DIC (PRADIC) that is developed by Idaho National Laboratory (INL).
An Integrated Risk Assessment Process of Safety-Related Digital I&C Systems in Nuclear Power Plants
Existing analyses of CCFs in I&C systems mainly focus on hardware failures.
On the Surrogate Gap between Contrastive and Supervised Losses
Recent theoretical studies have attempted to explain the benefit of the large negative sample size by upper-bounding the downstream classification loss with the contrastive loss.
Retrodiction beyond the Heisenberg uncertainty relation
In quantum mechanics, the Heisenberg uncertainty relation presents an ultimate limit to the precision by which one can predict the outcome of position and momentum measurements on a particle.
Quantum Physics
Pairwise Supervision Can Provably Elicit a Decision Boundary
A classifier built upon the representations is expected to perform well in downstream classification; however, little theory has been given in literature so far and thereby the relationship between similarity and classification has remained elusive.
Calibrated Surrogate Losses for Adversarially Robust Classification
Adversarially robust classification seeks a classifier that is insensitive to adversarial perturbations of test patterns.
Deep Learning Interfacial Momentum Closures in Coarse-Mesh CFD Two-Phase Flow Simulation Using Validation Data
Development of those closures traditionally rely on the experimental data and analytical derivation with simplified assumptions that usually cannot deliver a universal solution across a wide range of flow conditions.
Learning from Noisy Similar and Dissimilar Data
We perform a detailed investigation of this problem under two realistic noise models and propose two algorithms to learn from noisy S-D data.
Using Deep Learning to Explore Local Physical Similarity for Global-scale Bridging in Thermal-hydraulic Simulation
Current system thermal-hydraulic codes have limited credibility in simulating real plant conditions, especially when the geometry and boundary conditions are extrapolated beyond the range of test facilities.
Investigations of the Influences of a CNN's Receptive Field on Segmentation of Subnuclei of Bilateral Amygdalae
Segmentation of objects with various sizes is relatively less explored in medical imaging, and has been very challenging in computer vision tasks in general.
Computationally Efficient CFD Prediction of Bubbly Flow using Physics-Guided Deep Learning
To realize efficient computational fluid dynamics (CFD) prediction of two-phase flow, a multi-scale framework was proposed in this paper by applying a physics-guided data-driven approach.
A Semi-Supervised Framework for Automatic Pixel-Wise Breast Cancer Grading of Histological Images
The proposed EM approach is based on the collaborative filtering among the annotated and unannotated datasets.
Calibrated Surrogate Maximization of Linear-fractional Utility in Binary Classification
A clue to tackle their direct optimization is a calibrated surrogate utility, which is a tractable lower bound of the true utility function representing a given metric.
Classification from Pairwise Similarities/Dissimilarities and Unlabeled Data via Empirical Risk Minimization
In this paper, we derive an unbiased risk estimator which can handle all of similarities/dissimilarities and unlabeled data.
Imitation Learning from Imperfect Demonstration
Imitation learning (IL) aims to learn an optimal policy from demonstrations.
Unsupervised Domain Adaptation Based on Source-guided Discrepancy
A previously proposed discrepancy that does not use the source domain labels requires high computational cost to estimate and may lead to a loose generalization error bound in the target domain.
Classification from Pairwise Similarity and Unlabeled Data
Supervised learning needs a huge amount of labeled data, which can be a big bottleneck under the situation where there is a privacy concern or labeling cost is high.
Convex Formulation of Multiple Instance Learning from Positive and Unlabeled Bags
Multiple instance learning (MIL) is a variation of traditional supervised learning problems where data (referred to as bags) are composed of sub-elements (referred to as instances) and only bag labels are available.
