Generates realistic person images for re-id, overcoming pose variations.
problem Lack of cross-view paired training data and pose variations in person re-identification.
method Pose-normalization GAN (PN-GAN) for generating images conditioned on pose.
result Synthesized images enable learning invariant features free of pose variations.
Aerial robot estimates human pose and path using dynamic classifier selection.
problem Estimating human pose and trajectory from aerial video.
method Dynamic classifier selection architecture; perspective correction; HOG and CNN features; 64 pose-viewpoint classes.
result Dynamic classifier selection improves efficiency and accuracy.
In this paper, we deal with two challenges for measuring the similarity of the subject identities in practical video-based face recognition - the variation of the head pose in uncontrolled environments and the computational expense of processing videos. Since the frame-wise feature mean is unable to characterize the po…
Geometric Capsule Autoencoders group 3D points into parts and objects.
problem Learning object representations from 3D point clouds.
method Geometric capsules with pose and feature components, Multi-View Agreement voting mechanism.
result Learned representations enable object identification and canonical pose recovery.
Unsupervised mesh disentanglement separates identity and pose.
problem Geometric disentanglement for 3D deformable models.
method CFAN-VAE architecture using conformal factor and normal features.
result CFAN-VAE achieves state-of-the-art performance on unsupervised geometric disentanglement.
There are many forms of feature information present in video data. Principle among them are object identity information which is largely static across multiple video frames, and object pose and style information which continuously transforms from frame to frame. Most existing models confound these two types of represen…
Tensor neural network improves human pose classification from 3D skeleton data.
problem Efficiently processing spatiotemporal data for human pose classification.
method Proposes a tensor-based neural network with three components: spatiotemporal feature construction, tensor fusion, and tensor-based neural network processing.
result Achieves state-of-the-art performance in human pose classification.
Enhances visual localization using graph smoothing.
problem Inferring camera pose from a single image.
method Constructs a graph based on GPS coordinates and temporal information, then smooths feature representations.
result Significantly improves localization accuracy on large datasets.
Generative model creates realistic dance poses from music.
problem Generating human-like dance poses from music.
method Music feature encoder, pose generator, music genre classifier integrated.
result Generative autoregressive model synthesizes dance sequences up to 5,000 frames.
Computational approaches to drug discovery can reduce the time and cost associated with experimental assays and enable the screening of novel chemotypes. Structure-based drug design methods rely on scoring functions to rank and predict binding affinities and poses. The ever-expanding amount of protein-ligand binding an…
Study investigates how preprocessing, feature selection, and model selection affect performance on imbalanced genetic data.
problem Challenges in using machine learning on imbalanced genetic datasets.
method Comparative analysis of data preprocessing, feature selection techniques, and machine learning models on imbalanced genetic data.
result Class-imbalanced target variables and skewed predictors have little to no impact on classification performance.
Novel approach for forecasting financial time series using neural bag-of-features.
problem Forecasting financial time series with high-dimensionality, velocity, and variety.
method Temporal Logistic Neural Bag-of-Features combining deep neural networks.
result Demonstrated effectiveness on a large-scale financial time series dataset.
Scalable approach for object pose estimation across domains.
problem Object pose estimation across different datasets and models.
method Multi-path learning: shared encoder, object-specific decoders.
result Generalizes well from synthetic to real data and across various instances.
Paper proposes CNN with SIFT for rotation invariant feature extraction.
problem Max-pooling layer discards rotational information, leading to rotation invariance issues.
method Uses SIFT descriptor to capture orientation and spatial relationships.
result Improves feature extraction on MNIST and fashionMNIST datasets.
Paper optimizes estimation of quadratic functionals in nonparametric IV models.
problem Optimal estimation of a nonlinear functional in ill-posed inverse regression.
method Adaptive, minimax estimation using leave-one-out, sieve NPIV estimator with data-driven sieve dimension selection.
result Adaptive estimator achieves minimax optimal rate in various ill-posed cases.
Facial landmark localization and occlusion estimation for driver safety.
problem Robust facial landmark localization and occlusion estimation under harsh lighting and occlusion.
method Occluded Stacked Hourglass approach based on Stacked Hourglass network.
result State-of-the-art results in face detection, head pose, and occlusion estimation on various datasets.
New algorithm helps identify useful features in big data.
problem Challenges in selecting relevant features from high-dimensional data.
method HARVEST algorithm evaluates features in random subsets to identify potentially useful ones.
result Empirical analyses and simulations show the HARVEST algorithm is highly effective.
New Siamese Capsule Networks improve face verification in limited data settings.
problem Few-shot face verification with limited labeled data.
method Siamese Capsule Networks with contrastive loss and normalized pose features.
result Siamese Capsule Networks outperform baselines in few-shot face verification.
We consider the problem of learning object arrangements in a 3D scene. The key idea here is to learn how objects relate to human poses based on their affordances, ease of use and reachability. In contrast to modeling object-object relationships, modeling human-object relationships scales linearly in the number of objec…
MLS improves feature selection for imbalanced data.
problem Machine learning challenges with imbalanced high-dimensional data.
method Introduces Marginal Laplacian Score (MLS) for better feature selection.
result MLS improves performance on synthetic and public datasets.
Combines multiple data types to predict emotions in images.
problem Predicting emotions in images using various data types.
method Combines facial features, scene extraction, audio tonality, human pose, text-based tagging, and CNN predictions.
result Improves accuracy in emotion prediction compared to baseline methods.
Generative model disentangles 3D shapes into independent factors.
problem Learning rich representations of deformable 3D shapes.
method Supervised 3D mesh-convolutional Variational AutoEncoder with latent feature disentanglement.
result Explicit disentanglement of latent factors improves shape generation and downstream tasks.
We give two provably accurate feature-selection techniques for the linear SVM. The algorithms run in deterministic and randomized time respectively. Our algorithms can be used in an unsupervised or supervised setting. The supervised approach is based on sampling features from support vectors. We prove that the margin i…
Strategic feature manipulation helps learners identify meaningful variables in online regression settings.
problem Strategic feature manipulation poses a challenge for learners in online regression settings.
method Investigates how strategic feature manipulation by individuals can help learners recover meaningful features.
result Simple learner behavior can help accurately recover meaningful features and incentivize feature improvement.
This paper formulates and studies a general continuous-time behavioral portfolio selection model under Kahneman and Tversky's (cumulative) prospect theory, featuring S-shaped utility (value) functions and probability distortions. Unlike the conventional expected utility maximization model, such a behavioral model could…
Two new methods assess feature importance for fairness in machine learning models.
problem Understanding how features influence fairness in machine learning models.
method Two model-agnostic approaches: permutation and occlusion.
result Simple, scalable, and interpretable methods to quantify feature importance for fairness.
We study contextual linear bandit problems under feature uncertainty, where the features are noisy and have missing entries. To address the challenges posed by this noise, we analyze Bayesian oracles given the observed noisy features. Our Bayesian analysis reveals that the optimal hypothesis can significantly deviate f…
Greedy PIG adapts integrated gradients for better feature attribution.
problem Interpreting deep learning model predictions.
method Unified discrete optimization framework for feature attribution and selection.
result Greedy PIG improves feature attribution on various tasks.
New random feature maps for Laplacian and related kernels.
problem Challenges in approximating the Laplacian kernel and its generalizations.
method Developed random feature maps for Laplacian and related kernels, providing efficient sampling schemes.
result Demonstrated the efficacy of these random feature maps on real datasets.
Enhances GNNs with text features for better fake news detection.
problem Detecting disinformation on social media using GNNs.
method Integrates Transformer-based textual features into GNNs.
result Contextual text representations improve GNN performance by 33.8% in Macro F1.
Nonnegative matrix factorization (NMF) has become a very popular technique in machine learning because it automatically extracts meaningful features through a sparse and part-based representation. However, NMF has the drawback of being highly ill-posed, that is, there typically exist many different but equivalent facto…
Large-scale Hierarchical Classification (HC) involves datasets consisting of thousands of classes and millions of training instances with high-dimensional features posing several big data challenges. Feature selection that aims to select the subset of discriminant features is an effective strategy to deal with large-sc…
ReTaSA tackles continuous target shift in regression problems.
problem Continuous target shift in regression settings.
method Nonparametric regularized approach to estimate importance weight function.
result The method provides theoretical justification for the estimated importance weight function.
MISSION selects features for large datasets efficiently and interpretably.
problem Feature selection challenges in ultra large-scale datasets.
method MISSION uses Count-Sketch data structure for stochastic gradient descent.
result MISSION accurately and efficiently selects features on large-scale datasets.
A new visualization tool MD plot discovers interesting structures in continuous features.
problem Identifying interesting structures in data distributions, especially with skewed, clipped, or multimodal distributions.
method Proposes a new visualization tool called the mirrored density plot (MD plot) that does not require adjusting density estimation parameters.
result The MD plot outperforms conventional methods in identifying structures in complex distributions.
We propose to prune a random forest (RF) for resource-constrained prediction. We first construct a RF and then prune it to optimize expected feature cost & accuracy. We pose pruning RFs as a novel 0-1 integer program with linear constraints that encourages feature re-use. We establish total unimodularity of the constra…
Optimizes ligand binding poses using CNNs and atomic grids.
problem Improving the accuracy of docking predictions for drug discovery.
method Differentiable atomic grid representation, CNN for scoring and optimization.
result Iteratively-trained CNNs outperform single CNNs in optimizing poses.
Paper uses ML to predict stock price movements from order book data.
problem Forecasting stock price movements in financial markets.
method Combines handcrafted and ML features for three classifiers, evaluated on two setups.
result Machine Learning shows promise for this task, suggesting future research.
We present a method for synthesizing a frontal, neutral-expression image of a person's face given an input face photograph. This is achieved by learning to generate facial landmarks and textures from features extracted from a facial-recognition network. Unlike previous approaches, our encoding feature vector is largely…
The field of multiple view geometry has seen tremendous progress in reconstruction and calibration due to methods for extracting reliable point features and key developments in projective geometry. Point features, however, are not available in certain applications and result in unstructured point cloud reconstructions.…
Paper develops real-time through-wall human pose imaging using RF signals.
problem Developing real-time vision through opaque objects.
method Many-to-Many Encoder/Decoder Paradigm, student/teacher learning, Residual CNN, RPN, LSTM.
result Accurately predicts human pose through visual occlusion using RF signals.
Paper shows how to build fair classifiers with noisy sensitive features.
problem Fairness in learning with noisy sensitive features.
method Developed a method to adjust tolerance for fairness when sensitive features are noisy.
result It is possible to build fair classifiers with noisy sensitive features.
New method embeds RNN Seq2Seq models to visualize spatiotemporal data.
problem Visualizing and interpreting spatiotemporal data in sequence prediction tasks.
method Embedding approach to visualize and interpret RNN Seq2Seq model representations.
result Embedding space projections of RNN Seq2Seq models capture spatiotemporal dynamics.
Proposes an autoencoder-based unsupervised feature selection method.
problem Feature selection for high-dimensional data with non-linear structures.
method Autoencoder with spectral graph analysis for joint feature selection and manifold learning.
result Outperforms traditional methods in preserving both broad and local data structures.
This work tackles uncertainty quantification in tomography reconstruction.
problem Ill-posed nature of tomographic reconstruction leading to no unique solution.
method Gaussian process modeling to incorporate prior knowledge and experimental noises.
result Efficient uncertainty quantification in tomographic reconstruction.
Improved neural network predicts spectral functions more accurately than traditional methods.
problem Reconstructing real-time spectral functions from imaginary-time Green's functions is ill-posed and challenging.
method Feature Learning Network (FL-net) for enhanced prediction accuracy.
result FL-net achieves at least 20% improvement over traditional methods like MEM.
Contrastive learning properties studied, including feature suppression and hierarchical learning.
problem Feature suppression and hierarchical learning in contrastive learning.
method Generalized contrastive loss, instance-based contrastive learning, explicit and controllable competing features.
result Contrastive learning can suppress and prevent the learning of competing features.
Rare features reduce analysis effectiveness; new method aggregates them.
problem Reducing analysis effectiveness due to rare features in high-dimensional data.
method Aggregating rare features into denser features using a tree-based similarity encoding.
result Our method achieves high accuracy in predicting hotel ratings from reviews.