Recently, large language models (LLMs) have been widely used as recommender systems, owing to their reasoning capability and effectiveness in handling cold-start items. A common approach prompts an LLM with a target user's purchase history to recommend items from a candidate set, often enhanced with retrieval-augmented generation (RAG). Most existing RAG approaches retrieve purchase histories of users similar to the target user; however, these histories often contain noisy or weakly relevant information and provide little or no useful information for candidate items. To address these limitations, we propose ItemRAG, a novel RAG approach that shifts focus from coarse user-history retrieval to fine-grained item-level retrieval. ItemRAG augments the description of each item in the target user's history or the candidate set by retrieving items relevant to each. To retrieve items not merely semantically similar but informative for recommendation, ItemRAG leverages co-purchase information alongside semantic information. Especially, through their careful combination, ItemRAG prioritizes more informative retrievals and also benefits cold-start items. Through extensive experiments, we demonstrate that ItemRAG consistently outperforms existing RAG approaches under both standard and cold-start item recommendation settings.

Graph-based social recommendation (SocialRec) has emerged as a powerful extension of graph collaborative filtering (GCF), which leverages graph neural networks (GNNs) to capture multi-hop collaborative signals from user-item interactions. These methods enrich user representations by incorporating social network information into GCF, thereby integrating additional collaborative signals from social relations. However, existing GCF and graph-based SocialRec approaches face significant challenges: they incur high computational costs and suffer from limited scalability due to the large number of parameters required to assign explicit embeddings to all users and items. In this work, we propose PULSE (Parameter-efficient User representation Learning with Social Knowledge), a framework that addresses this limitation by constructing user representations from socially meaningful signals without creating an explicit learnable embedding for each user. PULSE reduces the parameter size by up to 50% compared to the most lightweight GCF baseline. Beyond parameter efficiency, our method achieves state-of-the-art performance, outperforming 13 GCF and graph-based social recommendation baselines across varying levels of interaction sparsity, from cold-start to highly active users, through a time- and memory-efficient modeling process.

Self-supervised learning (SSL) has emerged as a promising paradigm that presents supervisory signals to real-world problems, bypassing the extensive cost of manual labeling. Consequently, self-supervised anomaly detection (SSAD) has seen a recent surge of interest, since SSL is especially attractive for unsupervised tasks. However, recent works have reported that the choice of a data augmentation function has significant impact on the accuracy of SSAD, posing augmentation search as an essential but nontrivial problem due to lack of labeled validation data. In this paper, we introduce ST-SSAD, the first unsupervised approach to end-to-end augmentation tuning for SSAD. To this end, our work presents two key contributions. The first is a new unsupervised validation loss that quantifies the alignment between augmented training data and unlabeled validation data. The second is new differentiable augmentation functions, allowing data augmentation hyperparameter(s) to be tuned in an end-to-end manner. Experiments on two testbeds with semantic class anomalies and subtle industrial defects show that ST-SSAD gives significant performance gains over existing works. All our code and testbeds are available at https://github.com/jaeminyoo/ST-SSAD.

Time series anomaly detection (TSAD) finds many applications such as monitoring environmental sensors, industry KPIs, patient biomarkers, etc. A two-fold challenge for TSAD is a versatile and unsupervised model that can detect various different types of time series anomalies (spikes, discontinuities, trend shifts, etc.) without any labeled data. Modern neural networks have outstanding ability in modeling complex time series. Self-supervised models in particular tackle unsupervised TSAD by transforming the input via various augmentations to create pseudo anomalies for training. However, their performance is sensitive to the choice of augmentation, which is hard to choose in practice, while there exists no effort in the literature on data augmentation tuning for TSAD without labels. Our work aims to fill this gap. We introduce TSAP for TSA “on autoPilot”, which can (self-)tune augmentation hyperparameters end-to-end. It stands on two key components: a differentiable augmentation architecture and an unsupervised validation loss to effectively assess the alignment between augmentation type and anomaly type. Case studies show TSAP’s ability to effectively select the (discrete) augmentation type and associated (continuous) hyperparameters. In turn, it outperforms established baselines, including SOTA self-supervised models, on diverse TSAD tasks exhibiting different anomaly types.

Multi-criteria (MC) recommender systems, which utilize MC rating information for recommendation, are increasingly widespread in various e-commerce domains. However, the MC recommendation using training-based collaborative filtering, requiring consideration of multiple ratings compared to single-criterion counterparts, often poses practical challenges in achieving state-of-the-art performance along with scalable model training. To solve this problem, we propose CA-GF, a training-free MC recommendation method, which is built upon criteria-aware graph filtering for efficient yet accurate MC recommendations. Specifically, first, we construct an item-item similarity graph using an MC user-expansion graph. Next, we design CA-GF composed of the following key components, including 1) criterion-specific graph filtering where the optimal filter for each criterion is found using various types of polynomial low-pass filters and 2) criteria preference-infused aggregation where the smoothed signals from each criterion are aggregated. We demonstrate that CA-GF is (a) efficient: providing computational efficiency, offering an extremely fast runtime of less than 0.2 seconds even on the largest benchmark dataset, (b) accurate: outperforming benchmark MC recommendation methods, achieving substantial accuracy gains up to 24% compared to the best competitor, and (c) interpretable: providing interpretations for the contribution of each criterion to the model prediction based on visualizations.

Graphs are widely used for representing pairwise interactions in complex systems. Since such real-world graphs are large and often evergrowing, sampling subgraphs is useful for various purposes, including simulation, visualization, stream processing, representation learning, and crawling. However, many complex systems consist of group interactions (e.g., collaborations of researchers and discussions on online Q&A platforms) and thus are represented more naturally and accurately by hypergraphs than by ordinary graphs. Motivated by the prevalence of large-scale hypergraphs, we study the problem of sampling from real-world hypergraphs, aiming at answering (Q1) how can we measure the goodness of sub-hypergraphs, and (Q2) how can we efficiently find a “good” sub-hypergraph. Regarding Q1, we distinguish between two goals: (a) representative sampling , which aims at capturing the characteristics of the input hypergraph, and (b) back-in-time sampling , which aims at closely approximating a past snapshot of the input time-evolving hypergraph. To evaluate the similarity of the sampled sub-hypergraph to the target (i.e., the input hypergraph or its past snapshot), we consider 10 graph-level, hyperedge-level, and node-level statistics. Regarding Q2, we first conduct a thorough analysis of various intuitive approaches using 11 real-world hypergraphs. Then, based on this analysis, we propose MiDaS and MiDaS-B , designed for representative sampling and back-in-time sampling, respectively. Regarding representative sampling, we demonstrate through extensive experiments that MiDaS , which employs a sampling bias toward high-degree nodes in hyperedge selection, is (a) Representative : finding overall the most representative samples among 15 considered approaches, (b) Fast : several orders of magnitude faster than the strongest competitors, and (c) Automatic : automatically tuning the degree of sampling bias. Regarding back-in-time sampling, we demonstrate that MiDaS-B inherits the strengths of MiDaS despite an additional challenge—the unavailability of the target (i.e., past snapshot). It effectively handles this challenge by focusing on replicating universal evolutionary patterns, rather than directly replicating the target.

Self-supervised learning (SSL) is a growing torrent that has recently transformed machine learning and its many real world applications, by learning on massive amounts of unlabeled data via self-generated supervisory signals. Unsupervised anomaly detection (AD) has also capitalized on SSL, by self-generating pseudo-anomalies through various data augmentation functions or external data exposure. In this vision paper, we first underline the importance of the choice of SSL strategies on AD performance, by presenting evidences and studies from the AD literature. Equipped with the understanding that SSL incurs various hyperparameters (HPs) to carefully tune, we present recent developments on unsupervised model selection and augmentation tuning for SSL-based AD. We then highlight emerging challenges and future opportunities; on designing new pretext tasks and augmentation functions for different data modalities, creating novel model selection solutions for systematically tuning the SSL HPs, as well as on capitalizing on the potential of pretrained foundation models on AD through effective density estimation.

A hypergraph is a data structure composed of nodes and hyperedges, where each hyperedge is an any-sized subset of nodes. Due to the flexibility in hyperedge size, hypergraphs represent group interactions (e.g., co-authorship by more than two authors) more naturally and accurately than ordinary graphs. Interestingly, many real-world systems modeled as hypergraphs contain edge-dependent node labels, i.e., node labels that vary depending on hyperedges. For example, on co-authorship datasets, the same author (i.e., a node) can be the primary author in a paper (i.e., a hyperedge) but the corresponding author in another paper (i.e., another hyperedge).

Many real-world interactions (e.g., researcher collaborations and email communication) occur among multiple entities. These group interactions are naturally modeled as hypergraphs. In graphs, transitivity is helpful to understand the connections between node pairs sharing a neighbor, and it has extensive applications in various domains. Hypergraphs, an extension of graphs, are designed to represent group relations. However, to the best of our knowledge, there has been no examination regarding the transitivity of real-world group interactions. In this work, we investigate the transitivity of group interactions in real-world hypergraphs. We first suggest intuitive axioms as necessary characteristics of hypergraph transitivity measures. Then, we propose a principled hypergraph transitivity measure HyperTrans, which satisfies all the proposed axioms, with a fast computation algorithm Fast-HyperTrans. After that, we analyze the transitivity patterns in real-world hypergraphs distinguished from those in random hypergraphs. Lastly, we propose a scalable hypergraph generator THera. It reproduces the observed transitivity patterns by leveraging community structures, which are pervasive in real-world hypergraphs. Our code and datasets are available at https://github.com/kswoo97/hypertrans.

Group interactions are prevalent in various complex systems (e.g., collaborations of researchers and group discussions on online Q&A sites), and they are commonly modeled as hypergraphs. Hyperedges, which compose a hypergraph, are non-empty subsets of any number of nodes, and thus each hyperedge naturally represents a group interaction among entities. The higher-order nature of hypergraphs brings about unique structural properties that have not been considered in ordinary pairwise graphs.

How can we solve semi-supervised node classification in various graphs possibly with noisy features and structures? Graph neural networks (GNNs) have succeeded in many graph mining tasks, but their generalizability to various graph scenarios is limited due to the difficulty of training, hyperparameter tuning, and the selection of a model itself. Einstein said that we should "make everything as simple as possible, but not simpler." We rephrase it into the careful simplicity principle: a carefully-designed simple model can surpass sophisticated ones in real-world graphs. Based on the principle, we propose SlimG for semi-supervised node classification, which exhibits four desirable properties: It is (a) accurate, winning or tying on 10 out of 13 real-world datasets; (b) robust, being the only one that handles all scenarios of graph data (homophily, heterophily, random structure, noisy features, etc.); (c) fast and scalable, showing up to 18 times faster training in million-scale graphs; and (d) interpretable, thanks to the linearity and sparsity. We explain the success of SlimG through a systematic study of the designs of existing GNNs, sanity checks, and comprehensive ablation studies.

Given historical stock prices and sparse tweets, how can we accurately predict stock price movement? Many market analysts strive to use a large amount of information for stock price prediction, and Twitter is one of the richest sources of information presenting real-time opinions of people. However, previous works that use tweet data in stock movement prediction have suffered from two limitations. First, the number of tweets is heavily biased towards only a few popular stocks, and most stocks have insufficient evidence for accurate price prediction. Second, many tweets provide noisy information irrelevant of actual price movement, and extracting reliable information from tweets is as challenging as predicting stock prices.In this paper, we propose SLOT (Self-supervised Learning of Tweets for Capturing Multi-level Price Trends), an accurate method for stock movement prediction. SLOT has two main ideas to address the limitations of previous tweet-based models. First, SLOT learns embedding vectors of stocks and tweets in the same semantic space through self-supervised learning. The embeddings allow us to use all available tweets to improve the prediction for even unpopular stocks, addressing the sparsity problem. Second, SLOT learns multi-level relationships between stocks from tweets, rather than using them as direct evidence for prediction, making it robust to the unreliability of tweets. Extensive experiments on real world datasets show that SLOT provides the state-of-the-art accuracy of stock movement prediction.

Group interactions are prevalent in various complex systems (e.g., collaborations of researchers and group discussions on online Q&A sites), and they are commonly modeled as hypergraphs. Hyperedges, which compose a hypergraph, are non-empty subsets of any number of nodes, and thus each hyperedge naturally represents a group interaction among entities. The higher-order nature of hypergraphs brings about unique structural properties that have not been considered in ordinary pairwise graphs. In this tutorial, we offer a comprehensive overview of a new research topic called hypergraph mining. We first present recently revealed structural properties of real-world hypergraphs, including (a) static and dynamic patterns, (b) global and local patterns, and (c) connectivity and overlapping patterns. Together with the patterns, we describe advanced data mining tools used for their discovery. Lastly, we introduce simple yet realistic hypergraph generative models that provide an explanation of the structural properties.

Group interactions are prevalent in a variety of areas. Many of them, including email exchanges, chemical reactions, and bitcoin transactions, are directional, and thus they are naturally modeled as directed hypergraphs, where each hyperarc consists of the set of source nodes and the set of destination nodes. For directed graphs, which are a special case of directed hypergraphs, reciprocity has played a key role as a fundamental graph statistic in revealing organizing principles of graphs and in solving graph learning tasks. For general directed hypergraphs, however, even no systematic measure of reciprocity has been developed.In this work, we investigate the reciprocity of 11 real-world hypergraphs. To this end, we first introduce eight axioms that any reasonable measure of reciprocity should satisfy. Second, we propose HYPERREC, a principled measure of hypergraph reciprocity that satisfies all the axioms. Third, we develop FERRET, a fast and exact algorithm for computing the measure, whose search space is up to $10 ^{147} \times$ smaller than that of naive computation. Fourth, using them, we examine 11 real-world hypergraphs and discover patterns that distinguish them from random hypergraphs. Lastly, we propose REDI, an intuitive generative model for directed hypergraphs exhibiting the patterns. The code and the datasets are available at https://github.com/kswoo97/hyprec.

Given a graph with partial observations of node features, how can we estimate the missing features accurately? Feature estimation is a crucial problem for analyzing real-world graphs whose features are commonly missing during the data collection process. Accurate estimation not only provides diverse information of nodes but also supports the inference of graph neural networks that require the full observation of node features. However, designing an effective approach for estimating high-dimensional features is challenging, since it requires an estimator to have large representation power, increasing the risk of overfitting. In this work, we propose SVGA (Structured Variational Graph Autoencoder), an accurate method for feature estimation. SVGA applies strong regularization to the distribution of latent variables by structured variational inference, which models the prior of variables as Gaussian Markov random field based on the graph structure. As a result, SVGA combines the advantages of probabilistic inference and graph neural networks, achieving state-of-the-art performance in real datasets.

How can we effectively find the best structures in tree models? Tree models have been favored over complex black box models in domains where interpretability is crucial for making irreversible decisions. However, searching for a tree structure that gives the best balance between the performance and the interpretability remains a challenging task. In this paper, we propose Tart (Transition Matrix Representation with Transposed Convolutions), our novel generalized tree representation for optimal structural search. Tart represents a tree model with a series of transposed convolutions that boost the speed of inference by avoiding the creation of transition matrices. As a result, Tart allows one to search for the best tree structure with a few design parameters, achieving higher classification accuracy than those of baseline models in feature-based datasets.

Graphs are widely used for representing pairwise interactions in complex systems. Since such real-world graphs are large and often evergrowing, sampling a small representative subgraph is indispensable for various purposes: simulation, visualization, stream processing, representation learning, crawling, to name a few. However, many complex systems consist of group interactions (e.g., collaborations of researchers and discussions on online Q&A platforms), and thus they can be represented more naturally and accurately by hypergraphs (i.e., sets of sets) than by ordinary graphs.

Given a graph dataset, how can we augment it for accurate graph classification? Graph augmentation is an essential strategy to improve the performance of graph-based tasks, and has been widely utilized for analyzing web and social graphs. However, previous works for graph augmentation either a) involve the target model in the process of augmentation, losing the generalizability to other tasks, or b) rely on simple heuristics that lead to unreliable results. In this work, we introduce five desired properties for effective augmentation. Then, we propose NodeSam (Node Split and Merge) and SubMix (Subgraph Mix), two model-agnostic algorithms for graph augmentation that satisfy all desired properties with different motivations. NodeSam makes a balanced change of the graph structure to minimize the risk of semantic change, while SubMix mixes random subgraphs of multiple graphs to create rich soft labels combining the evidence for different classes. Our experiments on social networks and molecular graphs show that NodeSam and SubMix outperform existing approaches in graph classification.

How can we classify graph-structured data only with positive labels? Graph-based positive-unlabeled (PU) learning is to train a binary classifier given only the positive labels when the relationship between examples is given as a graph. The problem is of great importance for various tasks such as detecting malicious accounts in a social network, which are difficult to be modeled by supervised learning when the true negative labels are absent. Previous works for graph-based PU learning assume that the prior distribution of positive nodes is known in advance, which is not true in many real-world cases. In this work, we propose GRAB (Graph-based Risk minimization with iterAtive Belief propagation), a novel end-to-end approach for graph-based PU learning that requires no class prior. GRAB models a given graph as a Markov network and runs the marginalization and update steps iteratively. The marginalization step estimates the marginals of latent variables, while the update step trains a classifier network utilizing the computed priors in the objective function. Extensive experiments on five datasets show that GRAB achieves state-of-the-art accuracy, even compared with previous methods that are given the true prior.

How can we efficiently correlate multiple stocks for accurate stock movement prediction? Stock movement prediction has received growing interest in data mining and machine learning communities due to its substantial impact on financial markets. One way to improve the prediction accuracy is to utilize the correlations between multiple stocks, getting a reliable evidence regardless of the random noises of individual prices. However, it has been challenging to acquire accurate correlations between stocks because of their asymmetric and dynamic nature which is also influenced by the global movement of a market. In this work, we propose DTML (Data-axis Transformer with Multi-Level contexts), a novel approach for stock movement prediction that learns the correlations between stocks in an end-to-end way. DTML makes asymmetric and dynamic correlations by a) learning temporal correlations within each stock, b) generating multi-level contexts based on a global market context, and c) utilizing a transformer encoder for learning inter-stock correlations. DTML achieves the state-of-the-art accuracy on six datasets collected from various stock markets from US, China, Japan, and UK, making up to 13.8%p higher profits than the best competitors and the annualized return of 44.4% on investment simulation.

Given a multivariate time series, how can we forecast all of its variables efficiently and accurately? The multivariate forecasting, which is to predict the future observations of a multivariate time series, is a fundamental problem closely related to many real-world applications. However, previous multivariate models suffer from large model sizes due to the inefficiency of capturing complex intra-variable patterns and inter-variable correlations, resulting in poor accuracy. In this work, we propose AttnAR (attention-based autoregression), a novel approach for general multivariate forecasting which maximizes its model efficiency via separable structure. AttnAR first extracts variable-wise patterns by a mixed convolution extractor that efficiently combines deep convolution layers and shallow dense layers. Then, AttnAR aggregates the patterns by learning time-invariant attention maps between the target variables. AttnAR accomplishes the state-of-the-art forecasting accuracy in four datasets with up to 117.3 times fewer parameters than the best competitors.

How can we run graphical inference on large graphs efficiently and accurately? Many real-world networks are modeled as graphical models, and graphical inference is fundamental to understand the properties of those networks. In this work, we propose a novel approach for fast and accurate inference, which first samples a small subgraph and then runs inference over the subgraph instead of the given graph. This is done by the bounded treewidth (BTW) sampling, our novel algorithm that generates a subgraph with guaranteed bounded treewidth while retaining as many edges as possible. We first analyze the properties of BTW theoretically. Then, we evaluate our approach on node classification and compare it with the baseline which is to run loopy belief propagation (LBP) on the original graph. Our approach can be coupled with various inference algorithms: it shows higher accuracy up to 13.7% with the junction tree algorithm, and allows faster inference up to 23.8 times with LBP. We further compare BTW with previous graph sampling algorithms and show that it gives the best accuracy.

Given feature-based data, how can we accurately classify individual input and interpret the result of it? Ensemble models are often the best choice in terms of accuracy when dealing with feature-based datasets. However, interpreting the decision made by the ensemble model for individual input seems intractable. On the other hand, decision trees, although being prone to overfit, are considered as the most interpretable in terms of being able to trace the decision process of individual input. In this work, we propose Ensemble to Distilled Tree (EDiT), a novel distilling method that generates compact soft decision trees from ensemble models. EDiT exploits the interpretability of a tree-based structure by removing redundant branches and learning sparse weights, while enhancing accuracy by distilling the knowledge of ensemble models such as random forests (RF). Our experiments on eight datasets show that EDiT reduces the number of parameters of an RF by 6.4 to 498.4 times with a minor loss of classification accuracy.

Given graphs with millions or billions of vertices and edges, how can we efficiently make inferences based on partial knowledge? Loopy Belief Propagation(LBP) is a graph inference algorithm widely used in various applications including social network analysis, malware detection, recommendation, and image restoration. The algorithm calculates approximate marginal probabilities of vertices in a graph within a linear running time proportional to the number of edges. However, when it comes to real-world graphs with millions or billions of vertices and edges, this cost overwhelms the computing power of a single machine. Moreover, this kind of large-scale graphs does not fit into the memory of a single machine. Although several distributed LBP methods have been proposed, previous works do not consider the properties of real-world graphs, especially the effect of power-law degree distribution on LBP. Therefore, our work focuses on developing a fast and scalable LBP for such large real-world graphs on distributed environment.

Given an undirected network where some of the nodes are labeled, how can we classify the unlabeled nodes with high accuracy? Loopy Belief Propagation (LBP) is an inference algorithm widely used for this purpose with various applications including fraud detection, malware detection, web classification, and recommendation. However, previous methods based on LBP have problems in modeling complex structures of attributed networks because they manually and heuristically select the most important parameter, the propagation strength. In this paper, we propose Supervised Belief Propagation (SBP), a scalable and novel inference algorithm which automatically learns the optimal propagation strength by supervised learning. SBP is generally applicable to attributed networks including weighted and signed networks. Through extensive experiments, we demonstrate that SBP generalizes previous LBP-based methods and outperforms previous LBP and RWR based methods in real-world networks.