Counterfactual learning to rank (CLTR) relies on exposure-based inverse propensity scoring (IPS), a LTR-specific adaptation of IPS to correct for position bias. While IPS can provide unbiased and consistent estimates, it often suffers from high variance. Especially when little click data is available, this variance can cause CLTR to learn sub-optimal ranking behavior. Consequently, existing CLTR methods bring significant risks with them, as naively deploying their models can result in very negative user experiences. We introduce a novel risk-aware CLTR method with theoretical guarantees for safe deployment. We apply a novel exposure-based concept of risk regularization to IPS estimation for LTR. Our risk regularization penalizes the mismatch between the ranking behavior of a learned model and a given safe model. Thereby, it ensures that learned ranking models stay close to a trusted model, when there is high uncertainty in IPS estimation, which greatly reduces the risks during deployment. Our experimental results demonstrate the efficacy of our proposed method, which is effective at avoiding initial periods of bad performance when little data is available, while also maintaining high performance at convergence. For the CLTR field, our novel exposure-based risk minimization method enables practitioners to adopt CLTR methods in a safer manner that mitigates many of the risks attached to previous methods.
相關內容
Modern machine learning models deployed in the wild can encounter both covariate and semantic shifts, giving rise to the problems of out-of-distribution (OOD) generalization and OOD detection respectively. While both problems have received significant research attention lately, they have been pursued independently. This may not be surprising, since the two tasks have seemingly conflicting goals. This paper provides a new unified approach that is capable of simultaneously generalizing to covariate shifts while robustly detecting semantic shifts. We propose a margin-based learning framework that exploits freely available unlabeled data in the wild that captures the environmental test-time OOD distributions under both covariate and semantic shifts. We show both empirically and theoretically that the proposed margin constraint is the key to achieving both OOD generalization and detection. Extensive experiments show the superiority of our framework, outperforming competitive baselines that specialize in either OOD generalization or OOD detection. Code is publicly available at //github.com/deeplearning-wisc/scone.
Interpretable time series prediction is crucial for safety-critical areas such as healthcare and autonomous driving. Most existing methods focus on interpreting predictions by assigning important scores to segments of time series. In this paper, we take a different and more challenging route and aim at developing a self-interpretable model, dubbed Counterfactual Time Series (CounTS), which generates counterfactual and actionable explanations for time series predictions. Specifically, we formalize the problem of time series counterfactual explanations, establish associated evaluation protocols, and propose a variational Bayesian deep learning model equipped with counterfactual inference capability of time series abduction, action, and prediction. Compared with state-of-the-art baselines, our self-interpretable model can generate better counterfactual explanations while maintaining comparable prediction accuracy.
This paper presents a Predictive Maneuver Planning with Deep Reinforcement Learning (PMP-DRL) model for maneuver planning. Traditional rule-based maneuver planning approaches often have to improve their abilities to handle the variabilities of real-world driving scenarios. By learning from its experience, a Reinforcement Learning (RL)-based driving agent can adapt to changing driving conditions and improve its performance over time. Our proposed approach combines a predictive model and an RL agent to plan for comfortable and safe maneuvers. The predictive model is trained using historical driving data to predict the future positions of other surrounding vehicles. The surrounding vehicles' past and predicted future positions are embedded in context-aware grid maps. At the same time, the RL agent learns to make maneuvers based on this spatio-temporal context information. Performance evaluation of PMP-DRL has been carried out using simulated environments generated from publicly available NGSIM US101 and I80 datasets. The training sequence shows the continuous improvement in the driving experiences. It shows that proposed PMP-DRL can learn the trade-off between safety and comfortability. The decisions generated by the recent imitation learning-based model are compared with the proposed PMP-DRL for unseen scenarios. The results clearly show that PMP-DRL can handle complex real-world scenarios and make better comfortable and safe maneuver decisions than rule-based and imitative models.
Ensemble methods such as bagging and random forests are ubiquitous in various fields, from finance to genomics. Despite their prevalence, the question of the efficient tuning of ensemble parameters has received relatively little attention. This paper introduces a cross-validation method, ECV (Extrapolated Cross-Validation), for tuning the ensemble and subsample sizes in randomized ensembles. Our method builds on two primary ingredients: initial estimators for small ensemble sizes using out-of-bag errors and a novel risk extrapolation technique that leverages the structure of prediction risk decomposition. By establishing uniform consistency of our risk extrapolation technique over ensemble and subsample sizes, we show that ECV yields $\delta$-optimal (with respect to the oracle-tuned risk) ensembles for squared prediction risk. Our theory accommodates general ensemble predictors, only requires mild moment assumptions, and allows for high-dimensional regimes where the feature dimension grows with the sample size. As a practical case study, we employ ECV to predict surface protein abundances from gene expressions in single-cell multiomics using random forests. In comparison to sample-split cross-validation and $K$-fold cross-validation, ECV achieves higher accuracy avoiding sample splitting. At the same time, its computational cost is considerably lower owing to the use of the risk extrapolation technique. Additional numerical results validate the finite-sample accuracy of ECV for several common ensemble predictors under a computational constraint on the maximum ensemble size.
The function or performance of a network is strongly dependent on its robustness, quantifying the ability of the network to continue functioning under perturbations. While a wide variety of robustness metrics have been proposed, they have their respective limitations. In this paper, we propose to use the forest index as a measure of network robustness, which overcomes the deficiencies of existing metrics. Using such a measure as an optimization criterion, we propose and study the problem of breaking down a network by attacking some key edges. We show that the objective function of the problem is monotonic but not submodular, which impose more challenging on the problem. We thus resort to greedy algorithms extended for non-submodular functions by iteratively deleting the most promising edges. We first propose a simple greedy algorithm with a proved bound for the approximation ratio and cubic-time complexity. To confront the computation challenge for large networks, we further propose an improved nearly-linear time greedy algorithm, which significantly speeds up the process for edge selection but sacrifices little accuracy. Extensive experimental results for a large set of real-world networks verify the effectiveness and efficiency of our algorithms, demonstrating that our algorithms outperform several baseline schemes.
We consider the Bayesian estimation of the parameters of a finite mixture model from independent order statistics arising from imperfect ranked set sampling designs. As a cost-effective method, ranked set sampling enables us to incorporate easily attainable characteristics, as ranking information, into data collection and Bayesian estimation. To handle the special structure of the ranked set samples, we develop a Bayesian estimation approach exploiting the Expectation-Maximization (EM) algorithm in estimating the ranking parameters and Metropolis within Gibbs Sampling to estimate the parameters of the underlying mixture model. Our findings show that the proposed RSS-based Bayesian estimation method outperforms the commonly used Bayesian counterpart using simple random sampling. The developed method is finally applied to estimate the bone disorder status of women aged 50 and older.
Learning to Rank (LTR) methods are vital in online economies, affecting users and item providers. Fairness in LTR models is crucial to allocate exposure proportionally to item relevance. The deterministic ranking model can lead to unfair exposure distribution when items with the same relevance receive slightly different scores. Stochastic LTR models, incorporating the Plackett-Luce (PL) model, address fairness issues but have limitations in computational cost and performance guarantees. To overcome these limitations, we propose FairLTR-RC, a novel post-hoc model-agnostic method. FairLTR-RC leverages a pretrained scoring function to create a stochastic LTR model, eliminating the need for expensive training. Furthermore, FairLTR-RC provides finite-sample guarantees on a user-specified utility using distribution-free risk control framework. By additionally incorporating the Thresholded PL (TPL) model, we are able to achieve an effective trade-off between utility and fairness. Experimental results on several benchmark datasets demonstrate that FairLTR-RC significantly improves fairness in widely-used deterministic LTR models while guaranteeing a specified level of utility.
Multi-stage ranking pipelines have become widely used strategies in modern recommender systems, where the final stage aims to return a ranked list of items that balances a number of requirements such as user preference, diversity, novelty etc. Linear scalarization is arguably the most widely used technique to merge multiple requirements into one optimization objective, by summing up the requirements with certain preference weights. Existing final-stage ranking methods often adopt a static model where the preference weights are determined during offline training and kept unchanged during online serving. Whenever a modification of the preference weights is needed, the model has to be re-trained, which is time and resources inefficient. Meanwhile, the most appropriate weights may vary greatly for different groups of targeting users or at different time periods (e.g., during holiday promotions). In this paper, we propose a framework called controllable multi-objective re-ranking (CMR) which incorporates a hypernetwork to generate parameters for a re-ranking model according to different preference weights. In this way, CMR is enabled to adapt the preference weights according to the environment changes in an online manner, without retraining the models. Moreover, we classify practical business-oriented tasks into four main categories and seamlessly incorporate them in a new proposed re-ranking model based on an Actor-Evaluator framework, which serves as a reliable real-world testbed for CMR. Offline experiments based on the dataset collected from Taobao App showed that CMR improved several popular re-ranking models by using them as underlying models. Online A/B tests also demonstrated the effectiveness and trustworthiness of CMR.
Machine learning plays a role in many deployed decision systems, often in ways that are difficult or impossible to understand by human stakeholders. Explaining, in a human-understandable way, the relationship between the input and output of machine learning models is essential to the development of trustworthy machine-learning-based systems. A burgeoning body of research seeks to define the goals and methods of explainability in machine learning. In this paper, we seek to review and categorize research on counterfactual explanations, a specific class of explanation that provides a link between what could have happened had input to a model been changed in a particular way. Modern approaches to counterfactual explainability in machine learning draw connections to the established legal doctrine in many countries, making them appealing to fielded systems in high-impact areas such as finance and healthcare. Thus, we design a rubric with desirable properties of counterfactual explanation algorithms and comprehensively evaluate all currently-proposed algorithms against that rubric. Our rubric provides easy comparison and comprehension of the advantages and disadvantages of different approaches and serves as an introduction to major research themes in this field. We also identify gaps and discuss promising research directions in the space of counterfactual explainability.
We propose a new method for event extraction (EE) task based on an imitation learning framework, specifically, inverse reinforcement learning (IRL) via generative adversarial network (GAN). The GAN estimates proper rewards according to the difference between the actions committed by the expert (or ground truth) and the agent among complicated states in the environment. EE task benefits from these dynamic rewards because instances and labels yield to various extents of difficulty and the gains are expected to be diverse -- e.g., an ambiguous but correctly detected trigger or argument should receive high gains -- while the traditional RL models usually neglect such differences and pay equal attention on all instances. Moreover, our experiments also demonstrate that the proposed framework outperforms state-of-the-art methods, without explicit feature engineering.
小貼士
登錄享
相關主題
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191