Improves online ranker evaluation using multi-dueling bandits.
problem Efficiently evaluating ranking algorithms from limited user feedback.
method Generalized dueling bandits model for simultaneous comparisons of multiple rankers.
result Orders of magnitude improvement in performance compared to state-of-the-art algorithms.
OLCS-Ranker improves peptide identification accuracy and speed on hard datasets.
problem Efficiently identifying peptides from MS/MS data, especially on hard datasets with many false positives.
method Cost-sensitive online learning model and iterative online learning algorithm.
result OLCS-Ranker outperforms existing methods in accuracy and speed on large datasets.
Perceptron is a classic online algorithm for learning a classification function. In this paper, we provide a novel extension of the perceptron algorithm to the learning to rank problem in information retrieval. We consider popular listwise performance measures such as Normalized Discounted Cumulative Gain (NDCG) and Av…
Two-stage recommender systems show better performance when components interact rather than operate independently.
problem Two-stage recommender systems are often treated as sums of their parts, ignoring interactions between components.
method Used synthetic and real-world data to demonstrate interactions between ranker and nominators. Derived a generalization lower bound and proposed a Mixture-of-Experts approach to learn optimal item pools.
result Independent nominator training can lead to performance on par with random recommendations, highlighting the importance of interactions.
Improves text-to-SQL models by selecting the best SQL query from beam output.
problem Simplifying database query writing for natural language questions.
method Discriminative re-ranker using BERT fine-tuned classifier.
result Achieved top 4 score on Spider leaderboard.
AMBER method selects features efficiently using autoencoders and model-based elimination.
problem Efficiently selecting relevant features for classification.
method Greedy backward elimination using a ranker model and autoencoders.
result AMBER outperforms other feature selection methods in classification accuracy.
A new method reduces variance in training early-stage rankers for large-scale search systems.
problem Training early-stage rankers for large-scale search systems is challenging due to exploding variance in policy gradient methods.
method Proposes credit-assigned policy gradient (CA-PG) to mitigate variance in training early-stage rankers.
result CA-PG significantly reduces variance in training early-stage rankers compared to vanilla policy gradient.
BubbleRank improves online search results using safe exploration.
problem Learning user preferences from scratch in online ranking is costly and risky.
method BubbleRank combines offline and online learning, starting with an initial base list and improving it online by exchanging items.
result BubbleRank achieves a graceful degradation of n-step regret with a good initial base list.
Two-stage recommender systems struggle with exploration, leading to linear regret.
problem Linear regret in two-stage recommender systems due to exploration issues.
method Proposed a method to synchronize exploration strategies between the ranker and nominators using LinUCB.
result Demonstrated the effectiveness of the proposed algorithm experimentally.
CRL improves recommendation systems by reducing distribution shift.
problem Offline metrics fail to predict online performance due to distribution shift in recommender systems.
method Proposes an information-theoretic disentanglement criterion and a variational lower bound for better generalisation under distribution shift.
result CRL variants deliver substantial online gains in listener engagement compared to baseline models.
A new method predicts stock ranking uncertainty to improve trading performance during regime shifts.
problem Ranking models fail during regime shifts, leading to suboptimal performance.
method Adapting DEUP to rankers, predicting rank displacement and uncertainty, and proposing a two-level deployment policy.
result The two-level deployment policy improves risk-adjusted performance and indicates DEUP adds value mainly as a tail-risk guard.
Algorithm ranks assets in fluctuating markets.
problem Ranking assets in nonstationary time series.
method Naive Bayes asset ranker that adjusts weights based on performance.
result Outperforms traditional methods and S&P 500 index.
Optimizes crowdsourced preference-based subjective evaluation with online learning.
problem Large-scale evaluation of generative media using crowdsourcing due to combinatorial explosion.
method Automatic optimization of pair combination selections and evaluation volumes with online learning.
result Optimizes evaluation by reducing pair combinations and allocating optimal evaluation volumes.
Unified RL meta-learning framework for few-shot optimization.
problem Few-shot learning optimization problems.
method Generic RL meta-learning framework that learns optimal optimization algorithms.
result Significantly improved performance on few-shot tasks.
Proposes standards for evaluating online machine learning methods in evolving data streams.
problem Difficulty in evaluating online machine learning methods under realistic conditions.
method Proposes comprehensive evaluation standards, performance measures, and evaluation strategies.
result Provides a new Python framework (float) for modular integration of libraries and custom code.
Paper addresses inconsistency between offline and online LTR performance.
problem Inconsistency between offline and online LTR performance in E-commerce.
method Proposes an evaluator-generator framework to maximize evaluator score using reinforcement learning.
result Significant improvement in Conversion Rate (CR) over existing models.
Develops a method to estimate optimal policy value in online learning.
problem Challenges in evaluating ongoing policies in online learning environments.
method Doubly Robust Interval Estimation (DREAM) method.
result Valid inference on online conditional mean estimator with asymptotically normal distribution.
Paper tackles policy selection with logged data and limited online interactions.
problem Safe evaluation and deployment of offline reinforcement learning policies.
method Active offline policy selection combining logged data with online interaction.
result Improves upon state-of-the-art OPE estimates and pure online policy evaluation.
Paper develops robust policy evaluation for reinforcement learning with outlier and heavy-tailed rewards.
problem Outlier contamination and heavy-tailed rewards in reinforcement learning.
method Develops a fully online robust policy evaluation procedure and efficient statistical inference.
result Establishes the Bahadur-type representation of the estimator and develops an online inference procedure.
Study evaluates human vs. machine review generation, finds human assessments correlate better with lexical overlaps.
problem Evaluating natural language generation models for online reviews is challenging and inconsistent.
method Compared human evaluators with various automated evaluation methods, including discriminative and word overlap metrics.
result Human evaluators do not correlate well with discriminative evaluators, but correlate better with lexical overlaps.
New method improves SACOBRA's performance on high-conditioning optimization problems.
problem High-conditioning optimization problems with expensive objective functions.
method Online whitening applied to SACOBRA in the black-box optimization paradigm.
result Online whitening reduces optimization error by a factor of 10 to 1e12 compared to plain SACOBRA.
Contextual bandit algorithms have become popular for online recommendation systems such as Digg, Yahoo! Buzz, and news recommendation in general. \emph{Offline} evaluation of the effectiveness of new algorithms in these applications is critical for protecting online user experiences but very challenging due to their "p…
New algorithm reduces online hyperparameter optimization costs.
problem High cost of evaluating validation examples in online HPO.
method Modeling online HPO as a time-varying Bayesian optimization problem, proposing a costly feedback setting.
result Cost-efficient GP-UCB algorithm reaches human expert-level performance.
New method uses bandit feedback to better evaluate recommender systems.
problem Traditional offline evaluation of recommender systems is inaccurate.
method Exploits bandit feedback to estimate online performance.
result Bandit feedback provides more accurate offline evaluation.
POLA adapts learning rates for online time series prediction.
problem Adapting to changing data distributions in dynamic environments.
method Adaptive learning rate regulation for recurrent neural networks.
result POLA outperforms other online prediction methods in real-world datasets.
The study evaluates 62 classifiers for detecting online toxic comments.
problem Identifying and classifying toxic online commentary.
method Systematic evaluation of 62 classifiers representing 19 algorithmic families on the Jigsaw dataset.
result Simple bad word lists are most predictive of offensive commentary.
Online media provides opportunities for marketers through which they can deliver effective brand messages to a wide range of audiences. Advertising technology platforms enable advertisers to reach their target audience by delivering ad impressions to online users in real time. In order to identify the best marketing me…
The paper shows how ignoring temporal context in recommender systems evaluation leads to false confidence, proposing a method to embed temporal context.
problem The discrepancy between offline and online recommender system performance evaluation.
method Proposes a training procedure to embed temporal context into recommender systems and validates its advantage using multi-objective optimization.
result Including temporal context in recommender systems evaluation can improve recall@20 by up to 20%.
New framework for evaluating ad auctions using stochastic modeling.
problem Challenges in evaluating deterministic ad auctions.
method Repurposed bid landscape model to approximate propensity scores, enabling robust OPE estimators.
result Remarkable alignment with online A/B test results, achieving 92% MDA in CTR prediction.
OEUVRE estimates online loss with constant time and memory, outperforming other methods.
problem Accurately estimating expected loss in online learning.
method Recursive evaluation of each sample on current and previous models, using algorithmic stability for updates.
result Consistency, convergence rates, and concentration bounds proved for OEUVRE.
We model and correct bias in sequential evaluation, improving ranking accuracy.
problem Sequential evaluation bias in online, irrevocable scoring.
method Modeling the rating process, posing as statistical inference, proposing an online algorithm.
result Near-linear time, online algorithm with guarantees in ranking metrics, information theoretically optimal.
Study evaluates AD methods for fraud detection in online credit card payments.
problem Fraud detection in online credit card payments using anomaly detection methods.
method Assessed several recent anomaly detection methods and compared them with standard supervised learning methods.
result LightGBM outperforms other methods but is more sensitive to distribution shifts.
New approach for distributed online optimization of non-convex losses with sublinear regret.
problem Regret evaluation and consensus in distributed, multi-agent systems with non-convex losses.
method Composite regret metric and consensus-based online normalized gradient (CONGD) approach for pseudo-convex losses; offline optimization oracle for general non-convex losses.
result First sublinear regret bound for general distributed online non-convex learning.
Study uses online bootstrap for RL inference, showing effectiveness.
problem Statistical inference for RL parameters in online settings.
method Online bootstrap method applied to TD and GTD algorithms in RL.
result Method is distributionally consistent for policy evaluation inference.
Paper designs an online algorithm for nonparametric Hawkes processes estimation.
problem Estimating triggering functions of multivariate Hawkes processes.
method NPOLE-MHP algorithm with online estimation and stability guarantees.
result NPOLE-MHP achieves O(1/T) regret and stability. Framework for precise recall control in spatial conflation tasks.
problem Precise recall control in large-scale spatial conflation tasks to avoid downstream analytics failures and excessive manual review.
method End-to-end framework using equigrid bounding-box filter, CSR representation, neural ranker, and inverse-variance weighted ensemble of threshold estimators.
result Achieves exact recall with sub-percent variance over tens of millions of geometry pairs, runs on a single TPU v3 core.
Paper applies NEAT for dynamic credit evaluation using streaming data.
problem Dynamic credit evaluation using streaming data.
method Neuroevolution of Augmenting Topologies (NEAT) with enhancements.
result NEAT effectively handles dynamic credit evaluation with streaming data.
New definition of regret for nonconvex online learning models.
problem Intractability of standard regret measures for nonconvex models.
method Introduced a local gradient based regret definition.
result Our definition provides more interpretable bounds for forecasting.
Adaptive algorithm for online evaluation of targeted audiences in advertising.
problem Determining the right match between advertising creatives and target audiences.
method Contextual bandit approach to address audience overlap and learn optimal display policies.
result The proposed method is more efficient than traditional split-testing methods.
A new gradient estimator for online optimization with two function evaluations.
problem Online optimization of convex and Lipschitz functions with noisy data.
method L1-randomization approach for gradient estimation.
result Compared or better guarantees than previous methods for canceling noise.
This work establishes always-valid risk bounds for online matrix completion.
problem Challenges in establishing always-valid concentration inequalities for online matrix completion.
method Combines non-asymptotic martingale concentration and regularized low-rank matrix regression.
result Establishes always-valid risk bound process for online matrix completion.
ORL tackles robust online learning from noisy data.
problem Learning from noisy data with outliers in large-scale settings.
method ORL approach for online robust learning, scalable and robust.
result Provable robustness and efficiency advantages demonstrated.
GLCB uses Gated Linear Networks for online contextual bandits.
problem Online learning in contextual bandits with uncertainty estimation.
method Gated Linear Networks (GLNs) for prediction and uncertainty estimation.
result GLCB outperforms state-of-the-art methods in online contextual bandits.
New algorithm for online collaborative filtering using linear bandits and alternating least squares.
problem Online collaborative filtering with item recommendations over time.
method Combines linear bandits and alternating least squares for matrix factorization.
result Superior performance in cumulative regret and average cumulative NDCG over state-of-the-art algorithms.
Optimizes particle filtering for non-stationary environments.
problem Tracking and adapting to non-stationary environments in online prediction.
method Formulated an efficient particle filtering method using online mirror descent algorithm.
result Achieves optimal particle efficiency in non-stationary environments.
Study predicts purchasing decisions of online food delivery customers.
problem Understanding and predicting consumer purchasing decisions in online food delivery.
method Used machine learning techniques including CART, C4.5, random forest, and rule-based classifiers to predict purchasing decisions.
result C4.5 decision tree model outperformed others with 91.67% accuracy.
We present an online approach to portfolio selection. The motivation is within the context of algorithmic trading, which demands fast and recursive updates of portfolio allocations, as new data arrives. In particular, we look at two online algorithms: Robust-Exponentially Weighted Least Squares (R-EWRLS) and a regulari…
Fast online Kernel SVM for big data with limited resources.
problem Efficiently training SVM on large datasets with limited computational resources.
method Split input space using LVQ, train SVM in clusters, limit support vectors.
result Achieves high accuracy with high throughput on large datasets.