Automatically infers high dynamic range illumination from a single indoor photo.
problem Predicting accurate indoor illumination from a single image.
method End-to-end deep neural network trained in three steps: lighting classifier, scene light localization, and fine-tuning for intensity prediction.
result Significantly outperforms previous methods in recovering high-quality HDR illumination.
Spatio-temporal data is intrinsically high dimensional, so unsupervised modeling is only feasible if we can exploit structure in the process. When the dynamics are local in both space and time, this structure can be exploited by splitting the global field into many lower-dimensional "light cones". We review light cone …
Predict missing and future data points in light curves using scalable Gaussian Processes.
problem Gappy time-series data from commercial cameras confound light curve prediction.
method MuyGPs, a scalable framework for hyperparameter estimation of Gaussian Processes using nearest neighbors sparsification and local cross-validation.
result MuyGPs enable accurate prediction of missing and future data points in light curves.
A new method trains lightweight neural networks using a more complex 'booster' network.
problem Training high-performing neural networks for real-time tasks like CTR prediction.
method Rocket launching framework: a cumbersome net guides the training of a lightweight net.
result Lightweight models achieve performance previously only possible with complex models.
Study finds nighttime lights correlate with Indian GDP growth.
problem Accurate forecasting of Indian economic growth.
method Examined relationship between GDP and nighttime lights using DMSP and VIIRS datasets.
result Nighttime lights correlate with Indian GDP growth.
Artificial neural networks reduce computational costs for predicting excitation energy transfer properties in light-harvesting systems.
problem Computational limitations in predicting excitation energy transfer properties in light-harvesting systems.
method Use of artificial neural networks to bypass the computational limitations of established techniques.
result Artificial neural networks predict transfer times and transfer efficiencies with similar or higher accuracy than frequently used approximate methods.
Recurrent CNNs improve image classification in low light conditions.
problem Poor performance of CNNs in noisy images.
method Added recurrent connections to CNN layers to enhance robustness.
result gruCNNs outperform cCNNs in low signal-to-noise ratio images.
New ML models improve VVLC channel characterization for vehicular OWC.
problem Inaccurate channel models for VVLC due to mobility effects.
method Machine learning (ML) models incorporating ambient light, turbulence, and reflection effects.
result ML models predict VVLC channel loss and CFR more accurately than existing methods.
A new distillation framework predicts stock trading volumes more accurately with less model size.
problem Predicting stock trading volumes using regression models without class correlations.
method Transformed regression model into a probabilistic forecasting model, matching distributions and correlational relationships.
result Framework achieves superior prediction accuracy with significantly smaller model size.
Optimizes UAV deployment for VLC-enabled UAVs considering illumination distribution.
problem Optimizing UAV deployment for VLC-enabled UAVs with illumination distribution consideration.
method Formulated as an optimization problem, solved using GRUs and Gaussian mixture model.
result Achieves up to 22.1% reduction in transmit power compared to conventional methods.
The thesis explores prediction games with different opponents and moves, revealing intrinsic barriers and efficient algorithms.
problem Understanding and designing efficient algorithms for prediction games with various opponents and move orders.
method Geometric insights into three types of prediction games: general learning task, prediction with expert advice, and online convex optimization.
result Revealed intrinsic barriers and developed computationally efficient learning algorithms with strong theoretical guarantees.
We derive a novel norm that corresponds to the tightest convex relaxation of sparsity combined with an ℓ2 penalty. We show that this new {\em k-support norm} provides a tighter relaxation than the elastic net and is thus a good replacement for the Lasso or the elastic net in sparse prediction problems. Through …
Using open source data, we observe the fascinating dynamics of nighttime light. Following a global economic regime shift, the planetary center of light can be seen moving eastwards at a pace of about 60 km per year. Introducing spatial light Gini coefficients, we find a universal pattern of human settlements across dif…
Deep learning and prior maps improve traffic light recognition for autonomous cars.
problem Recognizing traffic lights for autonomous cars in urban environments.
method Combining deep learning-based detection with prior maps for traffic light identification and state recognition.
result The proposed system correctly identified relevant traffic lights along predefined routes.
Optimizes ESN hyperparameters for time series datasets using Bayesian optimization.
problem Determining optimal ESN hyperparameters for efficient model training.
method Bayesian optimization to find hyperparameters that generalize across similar time series.
result Reduces the number of ESN models needed for a dataset while maintaining good performance.
Optimizes kernel density ratios for better predictions and information measures.
problem Improving accuracy of kernel density estimates for density ratios.
method Derives an optimal weight function using calculus of variations.
result Reduces bias in kernel density estimates, leading to improved prediction posteriors and information-theoretic measures.
Study subjective perception of low light restored images and develop an unsupervised QA model.
problem Lack of subjective QA for low light restored images and challenges in collecting human opinion scores.
method Create a dataset, conduct subjective QA study, develop self-supervised contrastive learning technique to extract features.
result Unsupervised NR QA model achieves state-of-the-art performance for low light restored images.
The study examines light-like points on constant mean curvature hypersurfaces in Lorentzian manifolds.
problem Characterizing light-like points on constant mean curvature hypersurfaces in Lorentzian manifolds.
method Analyzing the first and second fundamental forms, and the exterior derivative of the determinant function.
result If a light-like point is degenerate, the hypersurface contains a light-like geodesic segment.
Classifies surfaces with zero mean curvature in a light cone.
problem Classifying surfaces with zero mean curvature in a light cone.
method Examined geodesics and screw motions, used Weierstrass representations.
result Complete classification of ruled zero mean curvature surfaces.
Solves surface problem in 3D light cone.
problem Björling problem for zero mean curvature surfaces in the three-dimensional light cone.
method Solves the Björling problem for zero mean curvature surfaces in the three-dimensional light cone.
result Constructs and classifies all rotational zero mean curvature surfaces.
Study light ray transform on Lorentzian manifolds without conjugate points.
problem Recovering spacelike singularities from weighted light ray transforms.
method Fourier Integral Operator analysis and filtered back-projection.
result Recovery of spacelike singularities from weighted light ray transforms without conjugate points.
Reconstructing manifold structure from boundary light observations.
problem Reconstructing Lorentzian manifold structure from boundary light observations.
method Constructive proof using Snell's law for reflections at the boundary.
result Topological, differentiable, and conformal structure of subsets of sources uniquely determined.
Paper extends previous result on hypersurfaces with degenerate light-like points.
problem Characterizing hypersurfaces with degenerate light-like points in Lorentzian manifolds.
method Analyzes C3-differentiable hypersurfaces, extending previous C4-differentiability result. result Same conclusion holds for C3-differentiable hypersurfaces as for C4-differentiable ones. Dropout is one of the key techniques to prevent the learning from overfitting. It is explained that dropout works as a kind of modified L2 regularization. Here, we shed light on the dropout from Bayesian standpoint. Bayesian interpretation enables us to optimize the dropout rate, which is beneficial for learning of wei…
This paper studies trade-offs in private prediction methods.
problem Leakage of training data information in machine learning predictions.
method Private training and private prediction methods with trade-offs.
result Private training methods outperform private prediction methods in various settings.
The natural topological, differentiable and geometrical structures on the space of light rays of a given spacetime are discussed. The relation between the causality properties of the original spacetime and the natural structures on the space of light rays are stressed. Finally, a symplectic geometrical approach to the …
GANs improve building performance model accuracy by integrating occupant behaviors.
problem Discrepancies between design and operation performance in buildings.
method Generative Adversarial Networks (GANs) to learn mixture models combining existing BPMs with occupant behaviors.
result Augmented BPMs significantly outperform existing BPMs in achieving specified performance targets.
Study light ray transform in pseudo-Euclidean space, derive inversion formula, and prove stability.
problem Analyzing light ray transform in pseudo-Euclidean space.
method Investigate normal operator, derive inversion formula, analyze as Fourier Integral Operator.
result Derive an inversion formula and prove stability estimates.
This paper builds a machine learning model to predict credit defaults for unsecured lending.
problem High credit defaults and delinquency rates in unsecured lending due to imbalanced data.
method Employing machine learning techniques, particularly SMOTE for imbalanced data, and evaluating models like LGBM Classifier.
result LGBM Classifier model outperforms other models in predicting credit defaults.
The paper explores the injectivity of the light ray transform on Lorentzian manifolds.
problem Injectivity of the light ray transform on functions and tensors.
method Analyzes injectivity conditions for scalar and tensor fields on stationary and static Lorentzian manifolds.
result Injectivity of the light ray transform on functions and tensors is proven under specific conditions.
In this paper, we are concerned with light-like extremal surfaces in curved spacetimes. It is interesting to find that under a diffeomorphic transformation of variables, the light-like extremal surfaces can be described by a system of nonlinear geodesic equations. Particularly, we investigate the light-like extremal su…
Maps complex plane polynomials to light-like polygons in Einstein Universe.
problem Mapping between complex plane polynomials and light-like polygons.
method Constructs geometric homeomorphism between moduli spaces.
result Found minimal Lagrangian maps between ideal polygons.
In this paper we prove several multiplicity results of t-periodic light rays in conformally stationary spacetimes using the Fermat metric and the extensions of the classical theorems of Gromoll-Meyer and Bangert-Hingston to Finsler manifolds. Moreover, we exhibit some stationary spacetimes with a finite number of t…
Constructs all real analytic germs of zero mean curvature surfaces in Lorentz-Minkowski 3-space.
problem Analyzing surfaces with light-like points in Lorentz-Minkowski 3-space.
method Applying the Cauchy-Kovalevski theorem for partial differential equations.
result Surfaces with light-like points in Lorentz-Minkowski 3-space contain a light-like line when they do not change causal types.
Automated classification of astronomical light curves for LSST.
problem Handling massive astronomical data from LSST.
method Gradient boosting of decision trees, feature extraction and selection, augmentation.
result Achieved one of the top results in the PLAsTiCC challenge.
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.
Mitigates gender bias amplification in model predictions.
problem Gender bias amplification in model predictions.
method Posterior regularization to mitigate bias.
result Almost removes gender bias amplification in model predictions.
Accelerates pulsar light curve inference with learned representations and optimization.
problem Computational expense of Markov chain Monte Carlo methods for posterior inference.
method Combining U-Net latent representations with local simulator-guided optimization.
result 120x reduction in inference time (24 hours to 12 minutes) with accuracy preserved.
David Gabai proves a 4D light bulb theorem using smooth techniques.
problem Constructive smooth 4-manifold theory advancements.
method Innovative use of classical moves in a smooth construction.
result First new hands-on advance in constructive smooth 4-manifold theory in a long time.
This paper improves prediction intervals for heteroskedastic regression.
problem Adaptive prediction intervals for heteroskedastic regression.
method Normalized and Mondrian conformal prediction methods.
result Conditional validity of chosen conformal predictors related to data-generating assumptions.
We introduce large scale analogues of topological monotone and light maps, which we call coarsely monotone and coarsely light maps respectively. We show that these two classes of maps constitute a factorization system on the coarse category. We also show how coarsely monotone maps arise from a reflection in a similar w…
Extends post-prediction inference method for more accurate AI/ML data analysis.
problem Naively using AI/ML predictions as true observations leads to biased results.
method Extends Wang et al. method to relax assumptions and incorporate a scaling factor.
result Yields unbiased point estimates and proper coverage in simulations.
The paper studies hypersurface evolution in a light-cone and curvature flow.
problem Investigating the evolution of hypersurfaces in a light-cone.
method Exploring variational problems associated with hypersurfaces and curvature flow.
result Established perpetual existence and smooth convergence of curvature flow to a circle.
This paper uses deep reinforcement learning to optimize traffic light timing.
problem Inefficient traffic light control leads to long delays and energy waste.
method Deep reinforcement learning model using convolutional neural network and prioritized experience replay.
result The proposed model reduces waiting time compared to existing methods.
New method designs multilayer nanoparticles using AI.
problem Difficult to design multilayer nanoparticles by trial and error.
method Combines genetic algorithm and neural network for inverse design.
result Successfully designs multilayer nanoparticles efficiently.
Paper proposes evaluating clinical models under domain shifts for reliability.
problem Clinical models often fail to generalize to new environments due to domain shifts.
method Develops realistic scenarios based on disease landscapes for multi-label classification.
result Deep clinical models can fail to generalize in specific data regimes.
Injectivity result for light ray transform on Lorentzian manifolds.
problem Injectivity of light ray transform on Lorentzian manifolds.
method Explicit relationship between geodesic and magnetic vector fields.
result Injectivity up to natural obstruction under certain conditions.
New welfare-based fairness notions align with existing error rate balance and predictive parity.
problem Aligning fairness notions with welfare-based criteria.
method Discussing and establishing conditions for envy freeness and prejudice freeness.
result Envy freeness and prejudice freeness are equivalent to error rate balance and predictive parity.