SoQal reduces oracle label requests in active learning by up to 35%.
problem Exploiting unlabelled data in healthcare requires costly oracle labeling.
method Dynamic questioning strategy to minimize oracle label requests.
result SoQal reduces oracle label requests by up to 35%.
New method detects concept drifts with fewer labels.
problem Real-world data drifts over time, affecting model performance.
method Hierarchical Hypothesis Testing with Request-and-Reverify strategy.
result Significant reduction in label requests with improved performance.
Memory augmented neural networks improve active learning for one-shot predictions.
problem Scarcity and cost of labeled training data in deep architectures.
method Memory augmented neural networks and Class Margin Sampling (CMS) for reinforcement learning.
result The proposed method outperforms existing baselines in label predictions and reduces label requests.
Active learning is a type of sequential design for supervised machine learning, in which the learning algorithm sequentially requests the labels of selected instances from a large pool of unlabeled data points. The objective is to produce a classifier of relatively low risk, as measured under the 0-1 loss, ideally usin…
We study active learning where the labeler can not only return incorrect labels but also abstain from labeling. We consider different noise and abstention conditions of the labeler. We propose an algorithm which utilizes abstention responses, and analyze its statistical consistency and query complexity under fairly nat…
CAMS selects best pre-trained model for unlabeled data points.
problem Efficiently utilizing pre-trained models and unlabeled data.
method Contextual active model selection algorithm with two components: contextual model selection and active query.
result CAMS requires less than 10% labeling effort compared to existing methods, achieving similar or better accuracy.
Adaptive region-based active learning seeks labels for complex data.
problem Efficiently label complex datasets with minimal human effort.
method Adaptive region partitioning and active learning for distinct predictors.
result Substantial empirical benefits over existing methods.
Algorithm learns user's reward function from hypothetical behaviors.
problem Aligning agent behavior with unknown user objectives.
method Synthesizes hypothetical behaviors, asks user for rewards, trains neural network.
result Significantly outperforms prior methods in learning reward models.
DAL uses disentanglement for automatic labeling in GAN-based active learning.
problem Reducing human labeling in GAN-based active learning.
method DAL leverages disentanglement in InfoGAN to automatically label datapoints, deciding human labeling based on disagreement with InfoGAN labels and label correction.
result DAL achieves better performance than existing GAN-based active learning approaches on image classification tasks.
Active learning suffers from biased non-response, which this paper addresses.
problem Active learning's effectiveness is compromised by biased non-response in real-world contexts.
method Proposes a cost-based correction to the sampling strategy, UCB-EU, to mitigate the impact of biased non-response.
result UCB-EU successfully reduces the harm from labelling non-response in many settings.
Proposes candidate labeling for more efficient use of crowdsourced labels.
problem Inefficiency of standard crowd learning where annotators provide single labels.
method Allows annotators to provide multiple labels per instance.
result Candidate labeling extracts more knowledge from labelers than standard approach.
Active learning selects most informative unlabeled samples for labeling.
problem Efficiently label unlabeled data in applications with scarce labeled data.
method Formulated as open-set recognition, uses VNNs to identify uncertain samples.
result Achieved state-of-the-art results on MNIST, CIFAR-10, and CIFAR-100.
New minimax results show how target labels benefit under covariate-shift.
problem Understanding the relative benefits of source and target labeled data under covariate-shift.
method Developed new minimax results and showed how a semi-supervised procedure can adapt to unknown transfer-exponent γ.
result Target labels can dramatically improve classification in certain regimes of covariate-shift.
Few-shot learning improved with semi-supervised and active methods.
problem Few-shot classification with limited labeled data.
method Prototypical Networks features combined with K-means clustering guided by few labeled examples and user feedback. result Active adaptation through user feedback significantly improves performance.
Android and Facebook provide third-party applications with access to users' private data and the ability to perform potentially sensitive operations (e.g., post to a user's wall or place phone calls). As a security measure, these platforms restrict applications' privileges with permission systems: users must approve th…
BiLA uses variational Bayesian inference to aggregate noisy labels online.
problem Aggregating noisy labels from crowd workers in real-time.
method Variational Bayesian inference and stochastic optimization.
result BiLA reduces label error by at least 10-1.5% points.
We introduce a semi-supervised discrete choice model to calibrate discrete choice models when relatively few requests have both choice sets and stated preferences but the majority only have the choice sets. Two classic semi-supervised learning algorithms, the expectation maximization algorithm and the cluster-and-label…
In crowd labeling, a large amount of unlabeled data instances are outsourced to a crowd of workers. Workers will be paid for each label they provide, but the labeling requester usually has only a limited amount of the budget. Since data instances have different levels of labeling difficulty and workers have different r…
Crowdsourcing platforms provide marketplaces where task requesters can pay to get labels on their data. Such markets have emerged recently as popular venues for collecting annotations that are crucial in training machine learning models in various applications. However, as jobs are tedious and payments are low, errors …
FASE-AL uses active learning to reduce labeling costs for data streams.
problem Reduction of labeling costs for data stream classification.
method Combines Fast Adaptive Stacking of Ensembles (FASE) with active learning.
result Achieves high accuracy with minimal labeled data.
Proposes proactive bed requests to reduce ED boarding and patient wait times.
problem Reduces ED boarding and patient wait times by proactively requesting inpatient beds.
method Formulates as a Markov decision process, uses predictions of admission probability and time to disposition to guide bed requests.
result Proactive aggregate bed requests can reduce boarding times by 30-70% and length of stay by 6-15%.
SAPSAM trains CNNs on lung CTs with binary labels, improving CPA detection and localization.
problem Chronic Pulmonary Aspergillosis (CPA) detection and localization on CT scans using binary labels.
method Binary labels, average intensity projections, 2D RGB-like images, hierarchical CNN architectures.
result High classification accuracy, precise localization, predictive power of 2-year survival.
New active learning method for kernel selection improves efficiency and accuracy.
problem Real-world applications where acquiring true labels is costly or time-consuming.
method Active Multiple Kernel Learning (AMKL) with adaptive kernel selection (AMKL-AKS).
result AMKL-AKS achieves optimal sublinear regret and better performance with fewer labeled data.
SoQal uses selective oracle questioning to improve active learning of cardiac signals.
problem Active learning of cardiac signals is challenging due to limited labelled data and high annotation costs.
method Proposes a framework combining selective oracle questioning and Bayesian active learning by consistency.
result SoQal outperforms baseline methods in active learning of cardiac signals, even with noisy oracles.
Develops active learning method for linear optimization with margin-based criterion.
problem Optimizing decisions in linear optimization problems with limited labeled data.
method Smart Predict-then-Optimize (SPO) loss and margin-based active learning algorithm.
result Algorithm achieves significantly fewer labels than naive supervised learning, especially for minimizing SPO loss.
Federated Learning improves mobile data privacy by training classifiers without sharing raw data.
problem Privacy concerns in packet classification due to sensitive data sharing.
method Apply Federated Learning to mobile packet classification tasks, training models without raw data sharing.
result Demonstrated effectiveness of the approach in terms of performance, cost, and privacy.
We formalize and verify double auctions for multiple-quantity trades.
problem Matching multiple-quantity trade requests in double auctions.
method Formalized algorithms, correctness proofs, Coq proof assistant, verified OCaml and Haskell programs.
result Automatic detection of violations in exchange systems.
Paper uses HodgeRank and information maximization for efficient crowdsourced ranking.
problem Crowdsourced ranking quality improvement with limited budget.
method Information maximization applied to HodgeRank for active sampling.
result Boosts sampling efficiency compared to traditional methods.
Parallel unlearning framework for inherited models reduces computational overhead.
problem Challenges in unlearning complex, evolving model networks.
method Chronologically Directed Acyclic Graph (DAG) and Fisher Inheritance Unlearning (FIUn) method.
result Significant reduction in computational overhead and efficient parallel unlearning.
Protocol minimizes disclosure in classification tasks.
problem Ensuring minimal disclosure in classification protocols.
method Developed a protocol for multi-party classification that minimizes non-responsive document disclosure.
result Guarantees minimal disclosure of non-responsive documents.
Video games improve vehicle routing performance.
problem Optimizing vehicle routes with unpredictable passenger requests.
method Replaced vehicle routing with a game, trained agents to play.
result General game-playing agents outperform traditional methods.
Paper tackles online facility location with user requests and provides a quasi-linear time algorithm.
problem Online facility location problem with user requests and site selection costs.
method Design of a novel quasi-linear time algorithm based on the Hedge algorithm.
result Good theoretical guarantees on the performance of the novel algorithm.
Optimizes resource allocation in a network with random job requests.
problem Minimizing costs while satisfying job requests within a budget.
method Formalizes as a repeated game, proposes an online saddle-point algorithm.
result Upper bounds for regret and constraint violations are derived.
Recognising human activities from streaming videos poses unique challenges to learning algorithms: predictive models need to be scalable, incrementally trainable, and must remain bounded in size even when the data stream is arbitrarily long. Furthermore, as parameter tuning is problematic in a streaming setting, suitab…
Predicts and classifies computational jobs for efficient resource allocation in cloud centers.
problem Efficiently scheduling and assigning resources to computational jobs in cloud centers.
method Applied LSTM neural network for job arrival prediction and BIRCH clustering for job classification.
result Improved accuracy in predicting and classifying computational jobs compared to existing methods.
The paper models CBF dynamics using queueing theory and insurance risk models.
problem Understanding and optimizing the operation of community bail funds.
method Combining queueing theory with classic insurance risk models.
result A fluid limit for the blocking model of CBF operations.
Teaches an active learner with contrastive examples to speed up learning.
problem Active learning with a helpful teacher and contrastive examples.
method Designs an adaptive teaching algorithm to provide informative contrastive examples.
result Strong performance guarantees and approximation guarantees for specific learners.
Graph autoencoders enable ML across diverse sensor networks.
problem Deploying ML across different sensor networks with varying types or layouts.
method Graph Autoencoders for activity recognition across heterogeneous sensor networks.
result Transferable activity classifiers achieve 75% accuracy on unseen sensor layouts.
MOVI learns optimal vehicle dispatch policies without models, reducing unserviced requests.
problem Optimizing vehicle dispatch to minimize passenger wait times in dynamic fleets.
method Model-free approach using Deep Q-network (DQN) for decentralized learning and centralized receding-horizon control comparison.
result DQN dispatch policy reduces unserviced requests by 76% compared to no dispatch and 20% compared to RHC.
I-AID categorizes disaster tweets into useful information types.
problem Filtering useful information from social media during disasters.
method Multimodel approach using BERT, GAT, and Relation Network.
result I-AID outperforms state-of-the-art approaches in F1 scores.
Active learning aims to obtain a classifier of high accuracy by using fewer label requests in comparison to passive learning by selecting effective queries. Many active learning methods have been developed in the past two decades, which sample queries based on informativeness or representativeness of unlabeled data poi…
The paper uses XAI to predict RFQ fulfillment accuracy.
problem Improving accuracy in predicting RFQ fulfillment for less liquid asset classes.
method Advanced algorithms like Logistic Regression, Random Forest, XGBoost, and Bayesian Neural Tree.
result Improved accuracy in RFQ fill rate predictions.
Conventional learning with expert advice methods assumes a learner is always receiving the outcome (e.g., class labels) of every incoming training instance at the end of each trial. In real applications, acquiring the outcome from oracle can be costly or time consuming. In this paper, we address a new problem of active…
This paper tackles delayed feedback in continuous training for CTR prediction, improving model performance by 3%.
problem Delayed feedback in CTR prediction leads to inferior performance and user experience.
method Comparing 5 loss functions and models in offline and online settings.
result Proposed methods outperform previous state-of-the-art by 3% relative cross entropy (RCE).
Market makers use a new method to predict and respond to RFQs in the OTC market.
problem Predicting and managing RFQs in the OTC market with Hawkes kernels.
method Developed a hierarchy of Volterra-Riccati approximations for path-dependent control problems.
result The state-feedback Volterra-Riccati policy closely tracks the exact benchmark and improves inventory and P&L risk control.
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.
In this paper we propose a novel framework for decentralized, online learning by many learners. At each moment of time, an instance characterized by a certain context may arrive to each learner; based on the context, the learner can select one of its own actions (which gives a reward and provides information) or reques…
New algorithm reduces costs and latency for large language model inference.
problem Optimizing inference costs and latency for large language models with GPU constraints.
method Formulated as an online scheduling problem with endogenous memory growth, introduced fluid model and WAIT algorithms.
result Reduced costs and latency, especially in near-overloaded and overloaded regimes.