New RL method learns from passive data by modeling intentions.
problem Learning from passive data like videos without rewards or actions.
method Model intentions using temporal difference learning, learning representations from raw data.
result Successfully learns features from passive data that accelerate downstream RL tasks.
Study allows removing data from machine learning models with strong guarantees.
problem Certifying removal of training data from machine learning models.
method Defined and developed a certified-removal mechanism for linear classifiers.
result Demonstrated that certified removal is possible and practical in certain learning settings.
VSAE learns from missing heterogeneous data by modeling latent dependencies.
problem Learning from partially-observed heterogeneous data with missingness.
method Variational selective autoencoder (VSAE) models joint distribution of observed, unobserved, and missing data.
result VSAE improves over state-of-the-art models in data generation and imputation tasks.
Library learns Bayesian networks from mixed data without discretization.
problem Learning Bayesian networks from mixed data (discrete and continuous variables).
method Proposes an algorithm for structural and parameter learning of Bayesian networks from mixed data using a mixed MI score function and Gaussian approximation. Offers two graph structure enumeration algorithms.
result Advantages in solving approximation and gap recovery problems on synthetic and real datasets.
Federated learning studies separate client data and distribution gaps.
problem Understanding performance differences in federated learning across different datasets.
method Proposed a framework to disentangle out-of-sample and participation gaps.
result Dataset synthesis strategy is crucial for realistic simulations of federated learning generalization.
MetaCL enables neural networks to learn from small data streams without forgetting.
problem Learning from limited data and adapting to new concepts over time.
method MetaCL trains a model to exploit intrinsic data features and dynamically penalize model parameter changes.
result MetaCL achieves state-of-the-art performance on image classification benchmarks.
Structured missing data complicates machine learning, presenting new challenges.
problem Structured missingness in data hinders machine learning at scale.
method No specific method is proposed; challenges are outlined.
result Structured missingness poses fundamental hindrance to machine learning.
Teaches manifolds from teacher's structured data.
problem Learning a manifold from teacher's structured data.
method Extends existing approaches to learning from randomly sampled data points, considering structured data provided by a teacher.
result Demonstrations can significantly reduce data points needed for manifold learning.
This paper proposes a generalization bound for GAN-synthetic data.
problem Improving classification accuracy and privacy in supervised learning.
method Proposes a generalization bound to measure the gap between synthetic and real data.
result Guarantees the generalization capability of classifiers learning from GAN-synthetic data.
SelectNet selectively adds unlabelled data to balance imbalanced training data.
problem Training from imbalanced data in machine learning.
method SelectNet trains a deep neural network to add unlabelled data with predicted labels to the training dataset.
result SelectNet effectively balances imbalanced training data, improving classifier performance.
Prototype learns automotive industry ontology from unstructured data.
problem Automatic learning of domain-specific ontologies from unstructured text data.
method Two-stage classification system: first classifier for concepts and irrelevant collocates, second classifier for concept types.
result Prototype validated with automotive industry complaint and repair data.
Paper tackles noisy S-D data for classification.
problem Learning from noisy Similar (S) and Dissimilar (D) pairs.
method Proposes two algorithms to learn from noisy S-D data under two noise models.
result Noise-informed algorithms outperform noise-blind baselines.
The paper tackles robust policy learning from multiple data sources.
problem Learning a policy that generalizes across diverse settings from multiple heterogeneous data sources.
method Proposes a minimax regret optimization objective and a policy learning algorithm combining doubly robust offline policy evaluation and no-regret learning.
result Achieves minimal worst-case mixture regret up to a moderated vanishing rate of the total data across all sources.
Paper tackles weakly supervised learning from similarity-confidence data.
problem Learning binary classifier from unlabeled data pairs with confidence of similarity.
method Proposes an unbiased estimator of classification risk from Sconf data and risk correction scheme.
result Demonstrates effectiveness of proposed methods through experiments.
Bayesian approach for learning from synthetic data, improving model accuracy.
problem Lack of statistical properties and robust methods for learning from synthetic data.
method Bayesian paradigm to update model parameters considering synthetic data generating process and learning task.
result Novel approach outperforms standard methods in supervised learning and inference problems.
New method learns from unreliable sources effectively.
problem Learning from untrusted, distributed, or private data sources.
method Statistical learning theory approach to suppress irrelevant or corrupted data.
result Significant improvements over alternative approaches in robust learning.
Survey of techniques for diagnosing pediatric sleep apnea from inexpensive data.
problem Diagnosing pediatric sleep apnea from limited and variable data.
method Exploratory data analysis using correlation networks, Mapper, SVD; supervised and unsupervised learning techniques.
result Analysis of various learning techniques applied to pediatric sleep apnea data.
Federated learning for data streams tackles real-time learning from IoT devices.
problem Efficiently learn from data streams generated by IoT devices and smartphones.
method Proposes a general federated learning algorithm for data streams using weighted empirical risk minimization.
result Demonstrates improved performance on various machine learning tasks compared to static dataset approaches.
DeGAN enriches data from related domains for future learning tasks.
problem Lack of relevant data for future learning tasks like Model Compression and Incremental Learning.
method Data-Enriching GAN (DeGAN) framework to retrieve representative samples from a trained classifier.
result State-of-the-art performance for Data-free Knowledge Distillation and Incremental Learning on benchmark datasets.
Survey on data collection challenges in machine learning.
problem Data scarcity and need for labeled data in machine learning.
method Comprehensive study of data acquisition, labeling, and improvement techniques.
result Identification of research challenges in data collection.
Detects anomalies without training data using deep learning.
problem Detect anomalies in data without labeled training data.
method Inverse Generative Adversarial Network (GAN) for semi-supervised learning.
result Successfully detects anomalies in data without labeled training data.
ORDisCo learns from unlabeled data to improve semi-supervised continual learning.
problem Lack of effective use of unlabeled data in semi-supervised continual learning.
method Deep Online Replay with Discriminator Consistency (ORDisCo) that continually passes the learned data distribution to a classifier and selectively stabilizes discriminator parameters.
result Significant performance improvement on various semi-supervised learning benchmark datasets.
New algorithms learn from fixed data without exploration.
problem Learning from fixed data without additional exploration.
method Introduces batch-constrained reinforcement learning.
result First continuous control RL algorithm for fixed data.
Unified approach to learn interpretable concepts from data.
problem Building interpretable machine learning models and highly-performing foundation models.
method Relating causal representation learning and foundation models, defining concepts and proving their recoverability.
result Provable recovery of human-interpretable concepts from diverse data.
A deep learning framework discovers causal relationships from incomplete data.
problem Discovering causal knowledge from incomplete observational data.
method Imputated Causal Learning (ICL) framework for iterative missing data imputation and causal structure discovery.
result ICL outperforms state-of-the-art methods in various missing data scenarios.
Develops machine learning models for clinical predictions from complex health data.
problem Complexity and uncertainty in medical data.
method Statistical Relational Learning approaches.
result Outstanding performance in medical research and real-world applications.
SAPAG attacks distributed learning by reconstructing true training data from gradients.
problem Privacy attacks on distributed learning systems through gradients.
method SAPAG uses a Gaussian kernel-based gradient difference distance measure.
result SAPAG can reconstruct training data on various DNNs and at different training phases.
Data Science is currently a popular field of science attracting expertise from very diverse backgrounds. Current learning practices need to acknowledge this and adapt to it. This paper summarises some experiences relating to such learning approaches from teaching a postgraduate Data Science module, and draws some learn…
Probabilistic methods improve SHM by learning from noisy, incomplete data.
problem Noisy and incomplete SHM data, lack of prior labels.
method Probabilistic algorithms for semi-supervised, active, and multi-task learning.
result Probabilistic methods enhance SHM by incorporating new data.
Picket guards against corrupted data in machine learning models.
problem Data corruption biases models and invalidates predictions.
method PicketNet detects corrupted data using self-supervised deep learning; flags corrupted queries online.
result Picket consistently protects models from corrupted data during training and deployment.
Machine learning predicts liquid water properties from cluster data.
problem Accuracy of bulk properties from machine-learned potentials is limited by training data.
method Local, atom-centred descriptors enable prediction of bulk properties from cluster data.
result Excellent agreement with experimental and theoretical counterparts of liquid water properties.
A central goal of unsupervised learning is to acquire representations from unlabeled data or experience that can be used for more effective learning of downstream tasks from modest amounts of labeled data. Many prior unsupervised learning works aim to do so by developing proxy objectives based on reconstruction, disent…
Study shows cliff-learning in transfer learning from foundation models.
problem Data-scaling of transfer learning from foundation models in low data regimes.
method Investigation of cliff-learning phenomenon through foundation-model analysis and toy models.
result Cliff-learning reflects compatibility between priors and tasks.
This research improves classification performance by learning a distance metric from balanced data.
problem Data imbalance in learning methods.
method Extracts a low-dimensional manifold, learns local neighborhood relationships, and optimizes distance metric.
result The proposed method outperforms other approaches, especially in imbalanced datasets.
Contrastive learning benefits from generated data but can be harmed by it too.
problem Contrastive learning's reliance on data augmentation and the impact of generated data.
method Investigates the role of generated data in contrastive learning and proposes Adaptive Inflation (AdaInf).
result Generated data can sometimes harm contrastive learning, and AdaInf improves performance.
New algorithms learn staged trees from incomplete data.
problem Handling missing data in staged tree learning.
method Characterized likelihood and pseudo-likelihoods for missing data, implemented structural expectation-maximization algorithm.
result It is feasible to account for different missingness patterns when learning staged trees.
A new method improves learning from imbalanced datasets by generating synthetic minority class samples.
problem Learning from imbalanced datasets where minority class is underrepresented.
method Clustering Based Oversampling: Generates synthetic data points based on the distance to cluster centroids.
result Improves learning from imbalanced data by incorporating the distribution structure of minority class samples.
Interpretable ML helps discover insights from big data.
problem Validating data-driven discoveries from complex datasets.
method Statistical and machine learning techniques for interpretable models.
result Challenges in validating data-driven discoveries remain.
Automates feature extraction from JSON data for machine learning.
problem Manual feature engineering for JSON data is laborious, lossy, and prone to bias.
method Automates feature extraction using Mill.jl and JsonGrinder.jl.
result Creates a differentiable machine learning model from raw JSON samples.
Meta-learning helps use small data from many tasks to compensate for lack of big data.
problem How to leverage small labeled data from many tasks to improve learning when big labeled data is scarce.
method Introduced a novel spectral approach to efficiently utilize small data tasks with the help of medium data tasks.
result The total number of examples necessary with only small data tasks scales similarly as when big data tasks are available.
A deep learning approach for efficient multidimensional projections.
problem Computational inefficiency and stability issues in existing projection methods.
method Train a deep neural network on sample projections to infer new ones.
result Generates projections similar to learned ones, faster and more stable.
ORIL learns a reward function from unlabeled data to improve robot learning.
problem Leveraging unlabeled data for robot learning.
method ORIL learns a reward function from demonstrator and unlabeled trajectories, annotates data, and trains an agent via offline reinforcement learning.
result ORIL consistently outperforms BC agents on various robotic tasks.
GANs learn from incomplete data with missing data imputation.
problem Learning from incomplete data with GANs.
method Proposes a GAN framework with a complete data generator and a mask generator to model missing data.
result Demonstrates effective imputation of missing data using adversarial training.
Survey of PU learning methods for positive and unlabeled data.
problem Learning from only positive and unlabeled data.
method Analyzes various approaches to PU learning.
result Identifies seven key research questions in PU learning.
Learning from small data sets is critical in many practical applications where data collection is time consuming or expensive, e.g., robotics, animal experiments or drug design. Meta learning is one way to increase the data efficiency of learning algorithms by generalizing learned concepts from a set of training tasks …
FedDAG learns DAG structures from decentralized data.
problem Learning DAG structures from private decentralized data.
method Gradient-based framework with two-level structures.
result FedDAG learns DAG structure without direct data sharing.
Bayesian model learns cancer subtypes from diverse NGS data.
problem Overdispersed NGS count data and limited samples for specific cancer types.
method Bayesian Multi-Domain Learning (BMDL) model using hierarchical negative binomial factorization.
result BMDL achieves reproducible cancer subtyping without negative transfer effects.
BL learns interpretable optimization structures from data.
problem Learning interpretable optimization structures from data.
method BL parameterizes a compositional utility function from intrinsically interpretable modular blocks.
result BL supports architectures from single to hierarchical compositions, modeling hierarchical optimization structures.