Gen1S learns novel classes with 1-shot data using residual space and generative models.
problem Learning new classes with limited data in a growing dataset.
method Mapping embeddings to a residual space, using generative models to learn multi-modal distribution, and applying it as a structural prior.
result Consistent improvement over state-of-the-art methods in recognizing novel classes.
Plug-and-play multimodal controller improves class-conditional image generation.
problem Generating class-conditional images from user-specified labels.
method Introduces a `multimodal controller` to generate multimodal data without additional learning parameters.
result Multimodal controlled generative models produce higher quality class-conditional images and novel modalities.
A deep generative model learns from seen and unseen classes without explicit training data.
problem Overcoming zero-shot learning with unseen classes.
method Variational auto-encoder with class-specific multi-modal prior, iteratively generating and learning unseen data.
result Outperforms models trained only on seen classes and state-of-the-art methods.
SNS-GAN integrates class labels into generative models for images and time series.
problem Effective integration of class labels in generative models without network modifications.
method Embeds class conditions within the generator's noise space.
result Superior performance in time series generation compared to baseline models.
Meta-learning framework improves zero-shot learning for unseen classes.
problem Handling unseen classes with strong bias towards seen classes.
method Meta-learning with Wasserstein GAN to handle class bias and unseen samples.
result Significant improvements in ZSL and GZSL settings.
Generative framework tackles zero-shot learning with adversarial domain adaptation.
problem Domain shift between seen and unseen class distributions in zero-shot learning.
method End-to-end learning of seen and unseen class distributions, adversarial domain adaptation.
result Superior accuracies compared to state-of-the-art models on various benchmark datasets.
A new model for latent class analysis with weighted responses.
problem Limitation of latent class model for real-world data with continuous or negative responses.
method Proposed a novel generative model, the weighted latent class model (WLCM).
result The proposed WLCM is more realistic and general than the latent class model.
C-t3VAE improves class representation in long-tailed generative models.
problem Latent geometric bias in VAEs under class imbalance.
method Per-class Student's t-distribution priors, closed-form objective, equal-weight latent mixture.
result Consistently lower FID scores and better class-balanced generation for severely imbalanced datasets.
N-VAE separates class-specific and shared factors in data.
problem Multimodal distribution of generative factors due to class distinction.
method N-VAE model with class-conditioned and shared latent spaces.
result Model effectively disentangles class-dependent and shared factors.
We present a generative framework for generalized zero-shot learning where the training and test classes are not necessarily disjoint. Built upon a variational autoencoder based architecture, consisting of a probabilistic encoder and a probabilistic conditional decoder, our model can generate novel exemplars from seen/…
Proposes a multimodal deep generative model for semi-supervised learning with class imbalance.
problem Class imbalance in semi-supervised learning with partial supervision.
method Separate encoders for each modality, sharing latent variables, and using Student's t-distributions for prior, encoder, and decoder.
result Outperforms baseline methods in generalization and classification performance for partially labeled multimodal data.
Analyzes how class imbalance and heterogeneity affect diffusion model learning dynamics.
problem Understanding how class imbalance and heterogeneity impact the learning dynamics of diffusion models.
method Developed a high-dimensional analytical framework to study class-dependent learning in score-based diffusion models.
result Class variance is the primary determinant of learning order, favoring higher-variance classes; centroid geometry plays a secondary role.
Generating user interpretable multi-class predictions in data rich environments with many classes and explanatory covariates is a daunting task. We introduce Diagonal Orthant Latent Dirichlet Allocation (DOLDA), a supervised topic model for multi-class classification that can handle both many classes as well as many co…
Adaptive model selection for RL with unknown function classes.
problem Model selection for RL with unknown function classes.
method Proposed adaptive algorithms that adapt to the smallest function class containing the true model.
result Cumulative regret matches that of an oracle with known function classes.
Graph-based framework for generalized few-shot learning.
problem Transferring learned models to novel tasks with few labeled examples.
method Graph-based framework that models relationships between seen and novel classes.
result Demonstrates benefits on benchmark datasets.
Generative model improves zero-shot sketch-based image retrieval.
problem Existing SBIR methods struggle with novel classes.
method Generative model learns to generate images conditioned on novel sketches.
result Significantly outperforms baselines on two challenging datasets.
TzK model learns from multiple datasets efficiently.
problem Efficiently learning from multiple heterogeneous datasets.
method Flow-based conditional generative model trained with maximum likelihood.
result Comparable log likelihood to state-of-the-art models.
CFA improves model's ability to generalize across unseen domain-class combinations.
problem Challenges in real-world machine learning applications due to data distribution shifts and limited training data.
method Developed Compositional Feature Alignment (CFA) technique to improve CG ability of pretrained models.
result CFA outperforms common finetuning techniques in compositional generalization.
Introduces model class selection to find sets of near-optimal models.
problem Finding sets of near-optimal models within multiple model collections.
method Generalizes model set selection framework to model class selection, using data splitting approaches.
result Shows that simpler, interpretable models can perform similarly to complex machine learning models.
Generative Adversarial Network model for class-imbalanced tabular data.
problem Class imbalance in binary classification problems.
method Generative Adversarial Network (GAN) with synthetic minority class samples.
result Improves average precision compared to re-weighting and oversampling techniques.
Confidence-based filtering reveals latent structure in diffusion models.
problem Unclear latent structure in diffusion models.
method Confidence scores from a classifier.
result Class-relevant latent structure emerges under confidence-based filtering.
Model for detecting rare labels in imbalanced crowdsourcing data.
problem Detecting rare labels in imbalanced crowdsourcing data.
method Generative aggregation model combining item difficulty and class-dependent annotator competence.
result Our model achieves the highest minority recall while maintaining competitive balanced accuracy.
CP-GAN generates images selectively conditioned on class specificity, capturing between-class relationships.
problem Generating images selectively conditioned on class specificity in class-overlapping data.
method Proposed Classifier's Posterior GAN (CP-GAN) that redesigns generator input and objective function for class-overlapping data.
result Demonstrated effectiveness of CP-GAN using both controlled and real-world class-overlapping data.
Generative models struggle with class prediction on real data.
problem Evaluating generative models' ability to infer class labels.
method Trained classifiers on synthetic data generated by various models and tested on real data.
result Generative models from different classes outperform GANs on a new classification accuracy score (CAS).
The paper tackles noisy data by focusing training on classes with high learnability.
problem Noisy labels in data collected via crowdsourcing or Web tagging.
method Develops an online algorithm that selects training data based on class learnability.
result The algorithm improves model generalization on learnable classes, leading to better performance.
Paper proposes a method to select base classes for few-shot learning.
problem How to select base classes for few-shot learning models.
method Formulated as a submodular optimization problem over Similarity Ratio.
result Our method effectively selects better base datasets for few-shot learning.
MCRAGE generates synthetic data to balance healthcare datasets.
problem Imbalanced datasets in healthcare lead to biased model performance for minority groups.
method Generative modeling to create synthetic data for underrepresented classes.
result MCRAGE improves model performance on minority groups.
New bounds study class-specific generalization error in machine learning.
problem Existing generalization theories assume uniform class performance, but in practice, classes vary significantly.
method Developed novel information-theoretic bounds using KL divergence and CMI.
result Theoretical bounds accurately capture complex class-generalization error behavior.
This report works out the details of a closed-form, fully Bayesian, multiclass, openset, generative pattern classifier using multivariate Gaussian likelihoods, with conjugate priors. The generative model has a common within-class covariance, which is proportional to the between-class covariance in the conjugate prior. …
Proposes a new framework for image generation using classification latent space representations.
problem Combining discriminative and dense representations for image generation and reconstruction.
method Discriminative modeling framework using manipulated supervised latent representations.
result Higher classification accuracy and visually realistic image generation compared to existing models.
MAIN network learns attributes without unseen class attributes for faster, more adaptable ZSL.
problem Learning unseen categories without known attributes and handling continual learning.
method Meta-learning attribute self-interaction network with inverse regularization.
result Main network outperforms state-of-the-art ZSL methods without unseen class attributes.
GANs improve VSR accuracy by generating unseen classes.
problem Scarcity of training data for unseen classes in VSR.
method Generative Adversarial Networks (GANs) to generate new class samples.
result Accuracy increased by 27% for unseen classes.
We present a simple generative framework for learning to predict previously unseen classes, based on estimating class-attribute-gated class-conditional distributions. We model each class-conditional distribution as an exponential family distribution and the parameters of the distribution of each seen/unseen class are d…
Regularization and data augmentation can be class-dependent, leading to poor performance on some classes.
problem Class-dependent effects of regularization and data augmentation.
method Evaluation of regularization and data augmentation techniques on Imagenet and INaturalist datasets.
result Regularization and data augmentation can lead to significant performance drops on some classes.
Improved model training for few-class, few-shot tasks.
problem Meta-learning algorithms struggle in many-shot and many-class settings.
method Joint training approach combining transfer-learning and meta-learning.
result Improved generalization performance on unseen tasks.
A new method for few-shot learning using embedded class models and shot-free meta training.
problem Few-shot learning with limited data and varying number of samples per class.
method Learning embeddings for few-shot learning with embedded class models and shot-free meta training.
result Achieves state-of-the-art performance on standard few-shot benchmark datasets.
A novel GAN framework improves multi-class classification by modeling label dependencies.
problem Improving multi-class classification performance through label dependency exploitation.
method Generative adversarial networks (GANs) where the discriminator learns label dependency and the generator learns to create dependent label sets.
result The discriminator significantly enhances generalization for various classification models.
ESRLCM clusters similar responses, more broadly than traditional models.
problem Clustering multivariate categorical data with common response patterns.
method Bayesian Equivalence Set Restricted Latent Class Model (ESRLCM).
result ESRLCM identifies clusters with similar item response probabilities.
Cost-effective method improves and re-purposes pre-trained GANs by fine-tuning class-embeddings.
problem Fine-tuning BigGANs from scratch is impractical due to instability and high computational cost.
method Fine-tuning only the class-embedding layer of pre-trained GANs.
result Significantly improved realism and diversity of samples, re-purposed for new tasks, and de-biased or improved diversity.
Optimal transport framework for zero-shot learning.
problem Generalized zero-shot learning of unseen classes.
method Conditional generative model and optimal transport between generated and real features.
result Optimal transport-based method outperforms state-of-the-art methods.
StepMix estimates mixture models with covariates for social science applications.
problem Estimating latent classes with covariates in social science models.
method Pseudo-likelihood estimation using one-, two-, and three-step approaches.
result Unified framework for expectation-maximization subroutines.
We demonstrate the usage of explicit form of the Thom class found by Mathai and Quillen for the definition of generating functional of a simple supersymmetric quantum mechanical model.
The paper studies generalization bounds for VRM, a variant of ERM.
problem Understanding the generalization performance of VRM.
method Proves generalization bounds for VRM under specific conditions.
result Generalization performance of VRM depends on vicinal function choice and function class quality.
Generative models characterized through learning theory.
problem Characterizing generative models using learning theory.
method Formalized Gold, Angluin, and Kleinberg's results; introduced uniform and non-uniform generation; characterized closure dimension.
result Incompatibility between generatability and predictability for certain hypothesis classes.
Generative replay extends sound classification models to new classes without old data.
problem Incrementally refining a sound classifier with new data causes previously learned tasks to degrade.
method Developed a generative replay procedure to generate training data in place of older datasets.
result Generative replay with 4% of old data performs as well as keeping 20% of old data.
Image classification datasets are often imbalanced, characteristic that negatively affects the accuracy of deep-learning classifiers. In this work we propose balancing GAN (BAGAN) as an augmentation tool to restore balance in imbalanced datasets. This is challenging because the few minority-class images may not be enou…
Predicts optimal training dataset sizes per class for machine learning models.
problem Optimizing training dataset sizes for class-specific machine learning models.
method Algorithm based on space-filling design of experiments, models like powerlaw curves and generalized linear models.
result The algorithm predicts optimal training dataset sizes per class for improved model performance.
Proposes DFDG for robust domain generalization without source domain labels.
problem Robustness of deep learning models in real-world applications where train and test distributions differ.
method Model-agnostic, class-aware alignment of class relationships through saliency maps.
result Competitive performance on time series sensor and image classification datasets.