Detecting anomalies in temporal data has gained significant attention across various real-world applications, aiming to identify unusual events and mitigate potential hazards. In practice, situations often involve a mix of segment-level labels (detected abnormal events with segments of time points) and unlabeled data (undetected events), while the ideal algorithmic outcome should be point-level predictions. Therefore, the huge label information gap between training data and targets makes the task challenging. In this study, we formulate the above imperfect information as noisy labels and propose NRdetector, a noise-resilient framework that incorporates confidence-based sample selection, robust segment-level learning, and data-centric point-level detection for multivariate time series anomaly detection. Particularly, to bridge the information gap between noisy segment-level labels and missing point-level labels, we develop a novel loss function that can effectively mitigate the label noise and consider the temporal features. It encourages the smoothness of consecutive points and the separability of points from segments with different labels. Extensive experiments on real-world multivariate time series datasets with 11 different evaluation metrics demonstrate that NRdetector consistently achieves robust results across multiple real-world datasets, outperforming various baselines adapted to operate in our setting.

Search engines are crucial in retrieving relevant items based on user-specified queries. A significant challenge arises when the buyer's vocabulary does not align with that of the seller, leading to a lack of sufficient recall or unsat-isfactory results. Such queries are referred to as “Null and Low” (N&L) queries which greatly hinder the overall user experience. Moreover, through analysis of user search behavioral data from a major e-commerce company, we have identified that approximately 29% of search queries exhibit multiple category interpretations, which we call “multi-faceted query interpretations”. In this study, we provide conceptual parallelism between the problem of N&L query reformulation and counterfactual explanation literature. To enhance the user experience for N&L queries, we propose a novel method that leverages the capabilities of a neural translation model to provide diverse and multiple reformulations. The proposed model demonstrated exceptional performance in our experiments, achieving an impressive 10% F-score improvement on the held-out test dataset with 5% improvement in relevance and a 100% increase in recall set size compared to a heuristic baseline, specifically for a set of N&L queries sampled from user traffic in eBay. By addressing the challenges of N&L queries and enabling the generation of diverse reformulations, our approach significantly enhances the overall search experience for users.

Personalized decision making requires the knowledge of potential outcomes under different treatments, and confidence intervals about the potential outcomes further enrich this decision-making process and improve its reliability in high-stakes scenarios. Predicting potential outcomes along with its uncertainty in a counterfactual world poses the foundamental challenge in causal inference. Existing methods that construct confidence intervals for counterfactuals either rely on the assumption of strong ignorability that completely ignores hidden confounders, or need access to un-identifiable lower and upper bounds that characterize the difference between observational and interventional distributions. In this paper, to overcome these limitations, we first propose a novel approach wTCP-DR based on transductive weighted conformal prediction, which provides confidence intervals for counterfactual outcomes with marginal converage guarantees, even under hidden confounding. With less restrictive assumptions, our approach requires access to a fraction of interventional data (from randomized controlled trials) to account for the covariate shift from observational distributoin to interventional distribution. Theoretical results explicitly demonstrate the conditions under which our algorithm is strictly advantageous to the naive method that only uses interventional data. Since transductive conformal prediction is notoriously costly, we propose wSCP-DR, a two-stage variant of wTCP-DR, based on split conformal prediction with same marginal coverage guarantees but at a significantly lower computational cost. After ensuring valid intervals on counterfactuals, it is straightforward to construct intervals for individual treatment effects (ITEs). We demonstrate our method across synthetic and real-world data, including recommendation systems, to verify the superiority of our methods compared against state-of-the-art baselines in terms of both coverage and efficiency. Our code can be found at https://github.com/rguo12/KDD24-Conformal.

Repeated risk minimization is a popular choice in real-world recommender systems driving their recommendation algorithms to adapt to user preferences and trends. However, numerous studies have shown that it exacerbates retention disparities among user groups, resulting in polarization within the user population. Given the primary objective of improving long-term user engagement in most industrial recommender systems and the significant commercial benefits from a diverse user population, enforcing retention fairness across user population is therefore crucial. Nonetheless, this goal is highly challenging due to the unknown dynamics of user retention (e.g., when a user would abandon the system) and the simultaneous aim to maximize the experience of every user.

The rawly collected training data often comes with separate noisy labels collected from multiple imperfect annotators (e.g., via crowdsourcing). A typical way of using these separate labels is to first aggregate them into one and apply standard training methods. The literature has also studied extensively on effective aggregation approaches. This paper revisits this choice and aims to provide an answer to the question of whether one should aggregate separate noisy labels into single ones or use them separately as given. We theoretically analyze the performance of both approaches under the empirical risk minimization framework for a number of popular loss functions, including the ones designed specifically for the problem of learning with noisy labels. Our theorems conclude that label separation is preferred over label aggregation when the noise rates are high, or the number of labelers/annotations is insufficient. Extensive empirical results validate our conclusions.

Recommendation systems aim to predict users' feedback on items not exposed to them yet. Confounding bias arises due to the presence of unmeasured variables (e.g., the socio-economic status of a user) that can affect both a user's exposure and feedback. Existing methods either (1) make untenable assumptions about these unmeasured variables or (2) directly infer latent confounders from users' exposure. However, they cannot guarantee the identification of counterfactual feedback, which can lead to biased predictions. In this work, we propose a novel method, i.e., identifiable deconfounder (iDCF), which leverages a set of proxy variables (e.g., observed user features) to resolve the aforementioned non-identification issue. The proposed iDCF is a general deconfounded recommendation framework that applies proximal causal inference to infer the unmeasured confounders and identify the counterfactual feedback with theoretical guarantees. Extensive experiments on various real-world and synthetic datasets verify the proposed method's effectiveness and robustness.

Crowdsourcing has been used to produce impactful and large-scale datasets for Machine Learning and Artificial Intelligence (AI), such as ImageNET, SuperGLUE, etc. Since the rise of crowdsourcing in early 2000s, the AI community has been studying its computational, system design, and data-centric aspects at various angles. We welcome the studies on developing and enhancing of crowdworker-centric tools, that offer task matching, requester assessment, instruction validation, among other topics. We are also interested in exploring methods that leverage the integration of crowdworkers to improve the recognition and performance of the machine learning models. Thus, we invite studies that focus on shipping active learning techniques, methods for joint learning from noisy data and from crowds, novel approaches for crowd-computer interaction, repetitive task automation, and role separation between humans and machines. Moreover, we invite works on designing and applying such techniques in various domains, including e-commerce and medicine.

This study aims to investigate robust algorithms for stochastic multi-armed bandit problems with adversarially corrupted rewards. We consider a novel setup of stochastic bandits where the corruptions are sporadic and adaptive to the learner’s arm selection strategy with no upper limit on the total budget constraint. We first introduce an attacker model called Fractional Oracle Attack (FOA), and show its efficacy against the standard UCB and ε-greedy algorithms with sufficient conditions for its success under $\mathcal{O}(\log T)$ attack cost. We then present two robust algorithms Sample Dropout-UCB (SD-UCB) and Sample Dropout-ε-greedy (SD-εG) to defend against FOA. The core idea of our algorithms is to use reward dropout during sample mean estimation, therefore tolerating a significant amount of quantified corruption. Both the algorithms are significantly more robust when compared to contemporary roust algorithms, and achieves a regret at the order of $\mathcal{O}(\log T)$.

''Knowledge should not be accessible only to those who can pay" said Robert May, chair of UC's faculty Academic Senate. Similarly, machine learning should not be accessible only to those who can pay. Thus, machine learning should benefit to the whole world, especially for developing countries in Africa and Asia. When dataset sizes grow bigger, it is laborious and expensive to obtain clean supervision, especially for developing countries. As a result, the volume of noisy supervision becomes enormous, e.g., web-scale image and speech data with noisy labels. However, standard machine learning assumes that the supervised information is fully clean and intact. Therefore, noisy data harms the performance of most of the standard learning algorithms, and sometimes even makes existing algorithms broken down.

Crowdsourcing enables the solicitation of forecasts on a variety of prediction tasks from distributed groups of people. How to aggregate the solicited forecasts, which may vary in quality, into an accurate final prediction remains a challenging yet critical question. Studies have found that weighing expert forecasts more in aggregation can improve the accuracy of the aggregated prediction. However, this approach usually requires access to the historical performance data of the forecasters, which are often not available. In this paper, we study the problem of aggregating forecasts without having historical performance data. We propose using peer prediction methods, a family of mechanisms initially designed to truthfully elicit private information in the absence of ground truth verification, to assess the expertise of forecasters, and then using this assessment to improve forecast aggregation. We evaluate our peer-prediction-aided aggregators on a diverse collection of 14 human forecast datasets. Compared with a variety of existing aggregators, our aggregators achieve a significant and consistent improvement on aggregation accuracy measured by the Brier score and the log score. Our results reveal the effectiveness of identifying experts to improve aggregation even without historical data.

In this paper we study the problem of how to detect and extract a particular type of propagation structure that arises in phishing activities. One of the most interesting phenomena induced by phishing is fast-flux, whereby a single malicious domain is mapped to a constantly changing IP address in order to evade capture and shut-down. This leads to malicious activities observed to be propagating through different networks, even though they originate from the same phishing campaign. To be able to detect and extract such a propagation is of significant importance as it can help us understand and analyze phishing activities. To achieve this goal, we propose a multi-layered propagation model, where layers correspond to different delay stages in the propagation and each is given by an adjacency matrix called the propagation matrix which models pairwise propagation relationships. A regression problem is then formulated to estimate this set of matrices so that the model prediction best fits the data; a Gibbs sampling based randomized algorithm is developed to efficiently find solutions with guaranteed performance. We evaluate our method using both simulation and Internet measurement data.