Paper proposes active learning for hotspot detection in VLSI design.
problem Hotspot detection in VLSI design is computationally expensive and relies on costly reference libraries.
method Active learning-based layout pattern sampling and hotspot detection flow.
result Significantly reduces lithography simulation overhead with satisfactory detection accuracy.
This work evaluates machine learning-based hotspot detectors on synthesized layout patterns.
problem Evaluating model robustness and generality of machine learning-based hotspot detectors.
method Developed an automatic layout generation tool to synthesize various layout patterns and tested machine learning-based detectors on these synthesized layouts.
result Machine learning-based detectors need continuous study for robustness and generality in DFM flows.
Paper presents a spatio-temporal Bayesian model for early detection of COVID-19 hotspots.
problem Understanding spatio-temporal dynamics of COVID-19 hotspots to prevent outbreaks.
method Spatio-temporal Bayesian framework with a zero-mean Gaussian process and non-stationary kernel function enhanced by deep neural networks.
result Model demonstrates superior hotspot-detection performance compared to baseline methods.
Adversarial perturbations can fool CNN-based lithographic hotspot detectors, but retraining can improve robustness.
problem Adversarial perturbations can mislead ML-based lithographic hotspot detectors.
method Proposed adversarial retraining strategy to improve robustness of CNN-based detectors.
result Adversarial retraining significantly improves robustness of CNN-based hotspot detection against perturbations.
Defense against DL-based lithographic hotspot detectors backdooring attacks reduces success rate from 84% to ~0%
problem DL-based lithographic hotspot detectors are vulnerable to backdoor attacks that can misclassify hotspots.
method Training data augmentation to eliminate intentional biases introduced during training.
result Significant reduction in attack success rate (from 84% to ~0%) using the proposed defense.
New method enhances hotspot prediction in IC designs.
problem Design hotspots vary between designs and are hard to predict.
method Synthetic pattern generation based on DOEs.
result Significantly reduces false alarms in hotspot prediction.
Survey of robust clustering methods for hotspot detection.
problem Detecting false positives in spatial hotspot mapping.
method Statistically rigorous clustering techniques.
result Survey of models and algorithms for robust clustering.
A new clustering algorithm tracks satellite hotspot data for bushfire tracking.
problem Tracking bushfires using satellite data.
method Spatiotemporal clustering algorithm with adjustable parameters.
result The algorithm effectively clusters satellite hotspot data for tracking bushfires.
Modeling infection hotspots to quantify effects of contact tracing and testing.
problem Capturing the role of infection hotspots in disease transmission.
method Temporal point process modeling framework to represent visits and disease transmission.
result Estimation of transmission rates at sites and households using Bayesian optimization.
Anomaly detection identifies unusual malaria transmission patterns in Ghana.
problem Identifying atypical malaria transmission patterns in Ghana's spatiotemporal surveillance data.
method Consensus-based anomaly detection framework applied to monthly malaria surveillance data.
result High-burden areas are not necessarily those with the most frequent anomalous transmission.
The first step in constructing a machine learning model is defining the features of the data set that can be used for optimal learning. In this work we discuss feature selection methods, which can be used to build better models, as well as achieve model interpretability. We applied these methods in the context of stres…
Robust model detects outliers in spatiotemporal epidemic data.
problem Outliers in epidemic data can mislead public health decisions.
method RST-GAM with mean-shift, adaptive Lasso, splines, and proximal algorithm.
result Demonstrates effectiveness in real-world COVID-19 data analysis.
Neural network ensembles predict design rule violations from early stages of IC design.
problem Predicting design rule violations from placement and global routing stages in IC design.
method Proposes a framework using neural network ensembles with soft voting and PCA-based subset selection.
result Significant improvement in model performance compared to baseline, including better performance than random forest.
Model predicts drug overdose hotspots using EMS and toxicology data.
problem Predicting drug overdose hotspots to focus limited services.
method Spatial-temporal point process model integrating EMS and toxicology data.
result Model improves prediction accuracy by integrating heterogeneous data.
Model predicts unseen climate extremes to inform risk planning.
problem Missing unseen climate extremes in historical records.
method DeepX-GAN model capturing spatial dependence.
result Unseen heat extremes disproportionately threaten vulnerable regions.
ATSDLN adapts to time series data for anomaly detection.
problem Challenges in selecting and optimizing anomaly detectors for time series data.
method Adaptive Time Series Detector Learning Network (ATSDLN) that selects and optimizes detectors and parameters.
result ATSDLN outperforms other methods in anomaly detection across various datasets.
Proposes a group-splicing algorithm for efficient BSGS in high-dimensional settings.
problem Efficiently selecting a small part of non-overlapping groups for best interpretability in high-dimensional settings.
method Iteratively detects relevant groups and excludes irrelevant ones using a novel group information criterion.
result Certifiable polynomial-time algorithm for identifying the optimal subset of groups with high probability.
In this study, the wind data series from five locations in Aegean Sea islands, the most active `hotspots' in terms of refugee influx during the Oct/2015 - Jan/2016 period, are investigated. The analysis of the three-per-site data series includes standard statistical analysis and parametric distributions, auto-correlati…
Optimizes e-hailing drivers' passenger seeking to reduce congestion and pollution.
problem Reduces congestion and pollution by optimizing e-hailing drivers' passenger seeking.
method Uses Markov Decision Process (MDP) and imitation learning to model and optimize drivers' decisions.
result Achieves a 17.5% improvement in passenger return rate over a heuristic strategy.
Automated rock fragmentation assessment using deep learning and spatial statistics.
problem Assessing post-blast rock fragmentation in real-time.
method Fine-tuned YOLO12l-seg model for instance segmentation, followed by spatial statistics.
result Framework accurately assesses rock fragmentation patterns in real-time.
Population migration is valuable information which leads to proper decision in urban-planning strategy, massive investment, and many other fields. For instance, inter-city migration is a posterior evidence to see if the government's constrain of population works, and inter-community immigration might be a prior evidenc…
DNN-based cross-modal retrieval has become a research hotspot, by which users can search results across various modalities like image and text. However, existing methods mainly focus on the pairwise correlation and reconstruction error of labeled data. They ignore the semantically similar and dissimilar constraints bet…
This paper reviews digital transformation research from 2011-2024, focusing on corporate finance.
problem Lack of systematic review in digital transformation from corporate finance perspective.
method Combines bibliometric and content analysis methods.
result Emerging and rapidly growing focus on digital transformation, particularly in developed countries.
Study estimates heterogeneous principal causal effects with binary treatments and intermediate variables.
problem Estimating subgroup effects within strata defined by potential values of an intermediate variable.
method Proposes a framework for estimating and forming confidence intervals for heterogeneous principal causal effects under principal ignorability assumption. Develops several estimators with varying robustness properties.
result Established large-sample theory and analyzed bias contributions of each approach.
Deep models can be fooled by maliciously crafted examples.
problem Deep learning models are vulnerable to adversarial examples that can mislead predictions without being noticed by humans.
method Reviews adversarial example generation methods and defenses, discusses their limitations and future prospects.
result Adversarial examples can mislead deep learning models without being distinguishable by humans.
This paper studies how social media posts, especially by executives, affect stock prices.
problem Predicting stock market movements using social media data.
method Integrated sentiment analysis of Twitter and Reddit posts with historical stock data using time series models and deep learning.
result Improvements in stock price prediction when social media data, especially executive posts, are included.
An explosion of high-throughput DNA sequencing in the past decade has led to a surge of interest in population-scale inference with whole-genome data. Recent work in population genetics has centered on designing inference methods for relatively simple model classes, and few scalable general-purpose inference techniques…
Study predicts wind energy potential in Gulf of Oman using climate models.
problem Predicting future wind energy potential in the Gulf of Oman.
method Used ERA5 and MENA simulations to project historical and future wind energy variability.
result Selected locations have suitable potential for wind power turbine construction.
Algorithm improves wildlife protection patrols.
problem Balancing exploration and exploitation in patrolling vast protected areas.
method Formulated as a stochastic multi-armed bandit problem, leveraging smoothness and decomposability.
result Algorithm LIZARD improves performance on real-world poaching data.
One pixel can significantly alter deep neural network outputs, revealing propagation patterns and vulnerability hotspots.
problem Understanding how a single pixel modification affects deep neural networks.
method Propagation Maps and locality analysis to visualize and understand the impact of pixel modifications.
result One pixel modifications can propagate through deep networks, affecting the final output and revealing vulnerability patterns.
Predictive policing models can be biased by differential crime reporting rates.
problem Bias in predictive policing models due to differential crime reporting.
method Simulation based on Bogotá, Colombia's victimization and crime reporting data.
result Differential crime reporting rates can lead to misallocation of police patrols.
Paper proposes methods to localize sources in WSNs without knowing sensor parameters.
problem Source localization in WSNs without sensor parameter knowledge.
method Hitting set approach and feature selection method.
result Effective source localization methods validated through simulations.
This paper compares deep learning and knowledge-based methods for pedestrian trajectory prediction.
problem Predicting pedestrian trajectories in crowded scenes is challenging due to external factors.
method Comprehensive comparison of deep learning and knowledge-based models.
result Deep learning models outperform knowledge-based models in local trajectory prediction.
AI enhances quantitative investment for better returns and risk control.
problem Achieving stable returns through AI in quantitative investment.
method Application of AI technology in quantitative investment strategies.
result AI improves investment performance and risk management.
This research predicts Bitcoin prices using wavelet and deep stacking approach.
problem Predicting price fluctuations of Bitcoin due to its price volatility.
method Wavelet for noise removal, deep learning models (neural networks and transformers), feature selection.
result The model achieved high accuracy in predicting Bitcoin prices at different time intervals.
Conditional diffusion models improve data generation with non-asymptotic convergence bounds.
problem Lack of non-asymptotic properties in conditional diffusion models.
method Integrates a pre-trained model into the diffusion model framework to capture conditional distributions.
result Established upper error bounds for the convergence between original and generated conditional distributions.
Deep Recurrent Survival Analysis models for better event prediction and survival rate estimation.
problem Survival analysis challenges in handling data censorship and sequential patterns.
method Combines deep learning for conditional probability prediction and survival analysis for censorship handling.
result Significantly outperforms state-of-the-art solutions in various metrics on real-world tasks.
Survey on machine learning from very few samples.
problem Learning and generalizing from very few samples.
method Comprehensive review of 300+ FSL papers.
result Meta learning based FSL approaches are emphasized.
New link detection results using closures of 3-braids.
problem Link detection using homology theories.
method Closure operations on 3-braids and homology theories.
result Detection of specific links using link Floer homology, Khovanov homology, and annular Khovanov homology.
Deep learning improves anomaly detection across various fields.
problem Detecting anomalies in data with advanced approaches.
method Survey of deep learning methods for anomaly detection.
result Advancements in deep anomaly detection address unique challenges.
Deep learning improves combustor anomaly detection in gas turbines.
problem Improving anomaly detection performance in gas turbine combustors.
method Hierarchically learned features from exhaust gas temperature sensor measurements using deep learning.
result Deep learning-based anomaly detection significantly improved combustor anomaly detection performance.
Reduced order modeling of energetic materials using physics-aware neural networks.
problem Simulating complex spatiotemporal dynamics in energetic materials.
method Physics-aware recurrent convolutions (PARC) combined with latent space projection to accelerate model training and inference.
result Significant decrease in training and inference time with comparable accuracy.
Paper proposes RL for real-time smart grid cyber attack detection.
problem Real-time detection of cyber-attacks in smart grids.
method Formulated as POMDP, uses model-free reinforcement learning.
result Effective in timely and accurate detection of cyber-attacks.
Graph energy helps detect communities in networks better than traditional methods.
problem Detecting communities in sparse networks where traditional methods fail.
method Using graph energy based on the full spectrum of adjacency matrices.
result The difference in graph energy between a planted partition model and an Erdős--Rényi network has a distinct transition at the detectability threshold.
New ML-based detection improves PMH signal detection in load-modulated MIMO systems.
problem Detecting PMH signals without prior CSI is challenging and computationally expensive.
method Proposes HEM-ML and HEM-KD schemes using EM and KD-tree for efficient detection.
result Achieves comparable detection results to optimal ML detector with reduced complexity.
Develops slope detection for 3-manifolds with torus boundaries.
problem Determining slopes on the boundary of 3-manifolds with torus boundaries.
method Introduces order-detection and representation-detection of slopes, proving their equivalence.
result Shows how slopes' behavior changes with cabling, improving previous results.
2DSig-Detect detects adversarial perturbations in images.
problem Adversarial attacks degrade image model performance.
method 2D-signature embedded semi-supervised framework using rough path theory.
result 2DSig-Detect outperforms other methods in detecting adversarial perturbations.
Develops a method to detect changes in linear systems with temporal correlations.
problem Detect abrupt changes in time series data with temporal correlations.
method Data-dependent threshold for online change point detection in linear dynamical systems.
result Achieves a pre-specified upper bound on the probability of false alarms and provides a finite-sample-based bound for detection probability.