Uncertainty quantification (UQ) plays a pivotal role in reduction of uncertainties during both optimization and decision making processes. It can be applied to solve a variety of real-world applications in science and engineering. Bayesian approximation and ensemble learning techniques are two most widely-used UQ methods in the literature. In this regard, researchers have proposed different UQ methods and examined their performance in a variety of applications such as computer vision (e.g., self-driving cars and object detection), image processing (e.g., image restoration), medical image analysis (e.g., medical image classification and segmentation), natural language processing (e.g., text classification, social media texts and recidivism risk-scoring), bioinformatics, etc. This study reviews recent advances in UQ methods used in deep learning. Moreover, we also investigate the application of these methods in reinforcement learning (RL). Then, we outline a few important applications of UQ methods. Finally, we briefly highlight the fundamental research challenges faced by UQ methods and discuss the future research directions in this field.

A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges

This paper proposes K-NRM, a kernel based neural model for document ranking. Given a query and a set of documents, K-NRM uses a translation matrix that models word-level similarities via word embeddings, a new kernel-pooling technique that uses kernels to extract multi-level soft match features, and a learning-to-rank layer that combines those features into the final ranking score. The whole model is trained end-to-end. The ranking layer learns desired feature patterns from the pairwise ranking loss. The kernels transfer the feature patterns into soft-match targets at each similarity level and enforce them on the translation matrix. The word embeddings are tuned accordingly so that they can produce the desired soft matches. Experiments on a commercial search engine's query log demonstrate the improvements of K-NRM over prior feature-based and neural-based states-of-the-art, and explain the source of K-NRM's advantage: Its kernel-guided embedding encodes a similarity metric tailored for matching query words to document words, and provides effective multi-level soft matches.

https://dl.acm.org/doi/pdf/10.1145/3077136.3080809

End-to-End Neural Ad-hoc Ranking with Kernel Pooling

With the rapid growth of web search in recent years the problem of modeling its users has started to attract more and more attention of the information retrieval community. This has several motivations. By building a model of user behavior we are essentially developing a better understanding of a user, which ultimately helps us to deliver a better search experience. A model of user behavior can also be used as a predictive device for non-observed items such as document relevance, which makes it useful for improving search result ranking. Finally, in many situations experimenting with real users is just infeasible and hence user simulations based on accurate models play an essential role in understanding the implications of algorithmic changes to search engine results or presentation changes to the search engine result page. In this survey we summarize advances in modeling user click behavior on a web search engine result page. We present simple click models as well as more complex models aimed at capturing non-trivial user behavior patterns on modern search engine result pages. We discuss how these models compare to each other, what challenges they have, and what ways there are to address these challenges. We also study the problem of evaluating click models and discuss the main applications of click models.

Click Models for Web Search

Learning to rank with biased click data is a well-known challenge. A variety of methods has been explored to debias click data for learning to rank such as click models, result interleaving and, more recently, the unbiased learning-to-rank framework based on inverse propensity weighting. Despite their differences, most existing studies separate the estimation of click bias (namely the propensity model ) from the learning of ranking algorithms. To estimate click propensities, they either conduct online result randomization, which can negatively affect the user experience, or offline parameter estimation, which has special requirements for click data and is optimized for objectives (e.g. click likelihood) that are not directly related to the ranking performance of the system. In this work, we address those problems by unifying the learning of propensity models and ranking models. We find that the problem of estimating a propensity model from click data is a dual problem of unbiased learning to rank. Based on this observation, we propose a Dual Learning Algorithm (DLA) that jointly learns an unbiased ranker and an unbiased propensity model. DLA is an automatic unbiased learning-to-rank framework as it directly learns unbiased ranking models from biased click data without any preprocessing. It can adapt to the change of bias distributions and is applicable to online learning. Our empirical experiments with synthetic and real-world data show that the models trained with DLA significantly outperformed the unbiased learning-to-rank algorithms based on result randomization and the models trained with relevance signals extracted by click models.

Unbiased Learning to Rank with Unbiased Propensity Estimation

Understanding user browsing behavior in web search is key to improving web search effectiveness. Many click models have been proposed to explain or predict user clicks on search engine results. They are based on the probabilistic graphical model (PGM) framework, in which user behavior is represented as a sequence of observable and hidden events. The PGM framework provides a mathematically solid way to reason about a set of events given some information about other events. But the structure of the dependencies between the events has to be set manually. Different click models use different hand-crafted sets of dependencies. We propose an alternative based on the idea of distributed representations: to represent the user's information need and the information available to the user with a vector state. The components of the vector state are learned to represent concepts that are useful for modeling user behavior. And user behavior is modeled as a sequence of vector states associated with a query session: the vector state is initialized with a query, and then iteratively updated based on information about interactions with the search engine results. This approach allows us to directly understand user browsing behavior from click-through data, i.e., without the need for a predefined set of rules as is customary for PGM-based click models. We illustrate our approach using a set of neural click models. Our experimental results show that the neural click model that uses the same training data as traditional PGM-based click models, has better performance on the click prediction task (i.e., predicting user click on search engine results) and the relevance prediction task (i.e., ranking documents by their relevance to a query). An analysis of the best performing neural click model shows that it learns similar concepts to those used in traditional click models, and that it also learns other concepts that cannot be designed manually.

https://staff.fnwi.uva.nl/m.derijke/wp-content/papercite-data/pdf/borisov-neural-2016.pdf

A Neural Click Model for Web Search

In modern search engines, an increasing number of search result pages (SERPs) are federated from multiple specialized search engines (called verticals, such as Image or Video). As an effective approach to interpret users' click-through behavior as feedback information, most click models were designed to reduce the position bias and improve ranking performance of ordinary search results, which have homogeneous appearances. However, when vertical results are combined with ordinary ones, significant differences in presentation may lead to user behavior biases and thus failure of state-of-the-art click models. With the help of a popular commercial search engine in China, we collected a large scale log data set which contains behavior information on both vertical and ordinary results. We also performed eye-tracking analysis to study user's real-world examining behavior. According these analysis, we found that different result appearances may cause different behavior biases both for vertical results (local effect) and for the whole result lists (global effect). These biases include: examine bias for vertical results (especially those with multimedia components), trust bias for result lists with vertical results, and a higher probability of result revisitation for vertical results. Based on these findings, a novel click model considering these biases besides position bias was constructed to describe interaction with SERPs containing verticals. Experimental results show that the new Vertical-aware Click Model (VCM) is better at interpreting user click behavior on federated searches in terms of both log-likelihood and perplexity than existing models.

/pdf/incorporating-vertical-results-into-search-click-models-2equi8w5ap.pdf

Incorporating vertical results into search click models

User's examination of search results is a key concept involved in all the click models. However, most studies assumed that eye fixation means examination and no further study has been carried out to better understand user's examination behavior. In this study, we design an experimental search engine to collect both the user's feedback on their examinations and the eye-tracking/click-through data. To our surprise, a large proportion (45.8%) of the results fixated by users are not recognized as being "read". Looking into the tracking data, we found that before the user actually "reads" the result, there is often a "skimming" step in which the user quickly looks at the result without reading it. We thus propose a two-stage examination model which composes of a first "from skimming to reading" stage (Stage 1) and a second "from reading to clicking" stage (Stage 2). We found that the biases (e.g. position bias, domain bias, attractiveness bias) considered in many studies impact in different ways in Stage 1 and Stage 2, which suggests that users make judgments according to different signals in different stages. We also show that the two-stage examination behaviors can be predicted with mouse movement behavior, which can be collected at large scale. Relevance estimation with the two-stage examination model also outperforms that with a single-stage examination model. This study shows that the user's examination of search results is a complex cognitive process that needs to be investigated in greater depth and this may have a significant impact on Web search.

/pdf/from-skimming-to-reading-a-two-stage-examination-model-for-3py2fbqgnf.pdf

From Skimming to Reading: A Two-stage Examination Model for Web Search

Click-through information is considered as a valuable source of users' implicit relevance feedback. As user behavior is usually influenced by a number of factors such as position, presentation style and site reputation, researchers have proposed a variety of assumptions (i.e.~click models) to generate a reasonable estimation of result relevance. The construction of click models usually follow some hypotheses. For example, most existing click models follow the sequential examination hypothesis in which users examine results from top to bottom in a linear fashion. While these click models have been successful, many recent studies showed that there is a large proportion of non-sequential browsing (both examination and click) behaviors in Web search, which the previous models fail to cope with. In this paper, we investigate the problem of properly incorporating non-sequential behavior into click models. We firstly carry out a laboratory eye-tracking study to analyze user's non-sequential examination behavior and then propose a novel click model named Partially Sequential Click Model (PSCM) that captures the practical behavior of users. We compare PSCM with a number of existing click models using two real-world search engine logs. Experimental results show that PSCM outperforms other click models in terms of both predicting click behavior (perplexity) and estimating result relevance (NDCG and user preference test). We also publicize the implementations of PSCM and related datasets for possible future comparison studies.

Incorporating Non-sequential Behavior into Click Models

While back-translation is simple and effective in exploiting abundant monolingual corpora to improve low-resource neural machine translation (NMT), the synthetic bilingual corpora generated by NMT models trained on limited authentic bilingual data are inevitably noisy. In this work, we propose to quantify the confidence of NMT model predictions based on model uncertainty. With word- and sentence-level confidence measures based on uncertainty, it is possible for back-translation to better cope with noise in synthetic bilingual corpora. Experiments on Chinese-English and English-German translation tasks show that uncertainty-based confidence estimation significantly improves the performance of back-translation.

Improving Back-Translation with Uncertainty-based Confidence Estimation

/pdf/improving-back-translation-with-uncertainty-based-confidence-2iap8j3122.pdf

Chao Wang

Papers

Incorporating vertical results into search click models

From Skimming to Reading: A Two-stage Examination Model for Web Search

Incorporating Non-sequential Behavior into Click Models

Improving Back-Translation with Uncertainty-based Confidence Estimation

Improving Back-Translation with Uncertainty-based Confidence Estimation