Graph Neural Networks (GNNs) have emerged as powerful tools for learning on graph-structured data, but often struggle to balance local and global information. While graph Transformers aim to address this by enabling long-range interactions, they often overlook the inherent locality and efficiency of Message Passing Neural Networks (MPNNs). We propose a new concept called 'fractal nodes', inspired by the fractal structure observed in real-world networks. Our approach is based on the intuition that graph partitioning naturally induces fractal structure, where subgraphs often reflect the connectivity patterns of the full graph. Fractal nodes are designed to coexist with the original nodes and adaptively aggregate subgraph-level feature representations, thereby enforcing feature similarity within each subgraph. We show that fractal nodes alleviate the over-squashing problem by providing direct shortcut connections that enable long-range propagation of subgraph-level representations. Experiment results show that our method improves the expressive power of MPNNs and achieves comparable or better performance to graph Transformers while maintaining the computational efficiency of MPNN by improving the long-range dependencies of MPNN.

Recently, data-driven simulators based on graph neural networks have gained attention in modeling physical systems on unstructured meshes. However, they struggle with long-range dependencies in fluid flows, particularly in refined mesh regions. This challenge, known as the 'over-squashing' problem, hinders information propagation. While existing graph rewiring methods address this issue to some extent, they only consider graph topology, overlooking the underlying physical phenomena. We propose Physics-Informed Ollivier--Ricci Flow (PIORF), a novel rewiring method that combines physical correlations with graph topology. PIORF uses Ollivier--Ricci curvature (ORC) to identify bottleneck regions and connects these areas with nodes in high-velocity gradient nodes, enabling long-range interactions and mitigating over-squashing. Our approach is computationally efficient in rewiring edges and can scale to larger simulations. Experimental results on 3 fluid dynamics benchmark datasets show that PIORF consistently outperforms baseline models and existing rewiring methods, achieving up to 26.2\% improvement.

Transformers have demonstrated remarkable performance across diverse domains. The key component of Transformers is self-attention, which learns the relationship between any two tokens in the input sequence. Recent studies have revealed that the self-attention can be understood as a normalized adjacency matrix of a graph. Notably, from the perspective of graph signal processing (GSP), the self-attention can be equivalently defined as a simple graph filter, applying GSP using the value vector as the signal. However, the self-attention is a graph filter defined with only the first order of the polynomial matrix, and acts as a low-pass filter preventing the effective leverage of various frequency information. Consequently, existing self-attention mechanisms are designed in a rather simplified manner. Therefore, we propose a novel method, called Attentive Graph Filter (AGF), interpreting the self-attention as learning the graph filter in the singular value domain from the perspective of graph signal processing for directed graphs with the linear complexity w.r.t. the input length. In our experiments, we demonstrate that AGF achieves state-of-the-art performance on various tasks, including Long Range Arena benchmark and time series classification. Code is available at https://github.com/hyowonwi/agf.

Recently, convolutional filters have been increasingly adopted in sequential recommendation for their ability to capture local sequential patterns. However, most of these models complement convolutional filters with self-attention. This is because convolutional filters alone, generally fixed filters, struggle to capture global interactions necessary for accurate recommendation. We propose \textbf{T}ime-\textbf{V}ariant Convolutional Filters for Sequential \textbf{Rec}ommendation (TV-Rec), a model inspired by graph signal processing, where time-variant graph filters capture position-dependent temporal variations in user sequences. By replacing both fixed kernels and self-attention with time-variant filters, TV-Rec achieves higher expressive power and better captures complex interaction patterns in user behavior. This design not only eliminates the need for self-attention but also reduces computation while accelerating inference. Extensive experiments on six public benchmarks show that TV-Rec outperforms state-of-the-art baselines by an average of 7.49\%.

The use of combination drugs among patients is increasing due to effectiveness compared to monotherapies. However, healthcare providers should continue to be concerned about the potential risks associated with patient safety arising from drug-drug interactions (DDIs) when they use combination drugs. Whereas direct physicochemical interactions contribute to certain cases of DDIs, the majority of DDIs occur because one drug modulates enzymes, such as cytochrome P450, responsible for metabolizing another drug. Therefore, drugs that interact with the same family drugs are more likely to interact with each other by mediating specific enzymes. Adapted from techniques used to recommend users with similar interests, we introduce an AI recommendation model with graph convolutional network (GCN) and collaborative filtering that analyzes the connectivity of interacting drugs rather than their chemical structures. This approach deviates from typical classification models by not requiring sampling of undefined interactions as negative samples, allowing the prediction of potential interactions for all unknown drug pairs, circumventing the challenges associated with selecting negative interactions and data imbalance. Our methodology used the DrugBank database (version 5.1.9 released on January 3, 2022), encompassing 4,072 drugs and 1,391,790 drug pairs with interactions. Furthermore, the robustness of the model was verified through a 5-fold validation and external data validation using TWOSIDES data. Notably, our model's efficacy is established solely through the exploitation of DDI reports, offering a versatile framework capable of accurately predicting interactions among diverse drug types. The source code for this project is distributed on GitHub (https://github.com/yeonuk-Jeong/DDI-OCF).

Sequential recommendation (SR) models based on Transformers have achieved remarkable successes. The self-attention mechanism of Transformers for computer vision and natural language processing suffers from the oversmoothing problem, i.e., hidden representations becoming similar to tokens. In the SR domain, we, for the first time, show that the same problem occurs. We present pioneering investigations that reveal the low-pass filtering nature of self-attention in the SR, which causes oversmoothing. To this end, we propose a novel method called Beyond Self-Attention for Sequential Recommendation (BSARec), which leverages the Fourier transform to i) inject an inductive bias by considering fine-grained sequential patterns and ii) integrate low and high-frequency information to mitigate oversmoothing. Our discovery shows significant advancements in the SR domain and is expected to bridge the gap for existing Transformer-based SR models. We test our proposed approach through extensive experiments on 6 benchmark datasets. The experimental results demonstrate that our model outperforms 7 baseline methods in terms of recommendation performance. Our code is available at https://github.com/yehjin-shin/BSARec.

Gross domestic product (GDP) nowcasting is crucial for policy-making as GDP growth is a key indicator of economic conditions. Dynamic factor models (DFMs) have been widely adopted by government agencies for GDP nowcasting due to their ability to handle irregular or missing macroeconomic indicators and their interpretability. However, DFMs face two main challenges: i) the lack of capturing economic uncertainties such as sudden recessions or booms, and ii) the limitation of capturing irregular dynamics from mixed-frequency data. To address these challenges, we introduce NCDENow, a novel GDP nowcasting framework that integrates neural controlled differential equations (NCDEs) with DFMs. This integration effectively handles the dynamics of irregular time series.NCDENow consists of 3 main modules: i) factor extraction leveraging DFM, ii) dynamic modeling using NCDE, and iii) GDP growth prediction through regression. We evaluate NCDENow against 6 baselines on 2 real-world GDP datasets from South Korea and the United Kingdom, demonstrating its enhanced predictive capability. Our empirical results favor our method, highlighting the significant potential of integrating NCDE into nowcasting models. Our code and dataset are available at https://github.com/sklim84/NCDENow_CIKM2024.

Recently, many mesh-based graph neural network (GNN) models have been proposed for modeling complex high-dimensional physical systems. Remarkable achievements have been made in significantly reducing the solving time compared to traditional numerical solvers. These methods are typically designed to i) reduce the computational cost in solving physical dynamics and/or ii) propose techniques to enhance the solution accuracy in fluid and rigid body dynamics. However, it remains under-explored whether they are effective in addressing the challenges of flexible body dynamics, where instantaneous collisions occur within a very short timeframe. In this paper, we present Hierarchical Contact Mesh Transformer (HCMT), which uses hierarchical mesh structures and can learn long-range dependencies (occurred by collisions) among spatially distant positions of a body --- two close positions in a higher-level mesh correspond to two distant positions in a lower-level mesh. HCMT enables long-range interactions, and the hierarchical mesh structure quickly propagates collision effects to faraway positions. To this end, it consists of a contact mesh Transformer and a hierarchical mesh Transformer (CMT and HMT, respectively). Lastly, we propose a flexible body dynamics dataset, consisting of trajectories that reflect experimental settings frequently used in the display industry for product designs. We also compare the performance of several baselines using well-known benchmark datasets. Our results show that HCMT provides significant performance improvements over existing methods. Our code is available at https://github.com/yuyudeep/hcmt.

Recent research in the field of graph neural network (GNN) has identified a critical issue known as "over-squashing," resulting from the bottleneck phenomenon in graph structures, which impedes the propagation of long-range information. Prior works have proposed a variety of graph rewiring concepts that aim at optimizing the spatial or spectral properties of graphs to promote the signal propagation. However, such approaches inevitably deteriorate the original graph topology, which may lead to a distortion of information flow. To address this, we introduce an ex**pand**ed width-**a**ware (**PANDA**) message passing, a new message passing paradigm where nodes with high centrality, a potential source of over-squashing, are selectively expanded in width to encapsulate the growing influx of signals from distant nodes. Experimental results show that our method outperforms existing rewiring methods, suggesting that selectively expanding the hidden state of nodes can be a compelling alternative to graph rewiring for addressing the over-squashing.

Structured data, which constitutes a significant portion of existing data types, has been a long-standing research topic in the field of machine learning. Various representation learning methods for tabular data have been proposed, ranging from encoder-decoder structures to Transformers. Among these, Transformer-based methods have achieved state-of-the-art performance not only in tabular data but also in various other fields, including computer vision and natural language processing. However, recent studies have revealed that self-attention, a key component of Transformers, can lead to an oversmoothing issue. We show that Transformers for tabular data also face this problem. To tackle the problem, we suggest a novel self-attention layer for tabular data, leveraging matrix polynomials. This proposed layer serves as a replacement for the original self-attention layer, contributing to the improvement of model scalability. In our experiments with three representative table learning models equipped with our proposed layer, we illustrate that the layer effectively mitigates the oversmoothing problem and enhances the representation performance of the existing methods, outperforming the state-of-the-art table representation methods.

Transformers, renowned for their self-attention mechanism, have achieved state-of-the-art performance across various tasks in natural language processing, computer vision, time-series modeling, etc. However, one of the challenges with deep Transformer models is the oversmoothing problem, where representations across layers converge to indistinguishable values, leading to significant performance degradation. We interpret the original self-attention as a simple graph filter and redesign it from a graph signal processing (GSP) perspective. We propose a graph-filter-based self-attention (GFSA) to learn a general yet effective one, whose complexity, however, is slightly larger than that of the original self-attention mechanism. We demonstrate that GFSA improves the performance of Transformers in various fields, including computer vision, natural language processing, graph-level tasks, speech recognition, and code classification.

With the rapid growth of cloud services driven by advancements in web service technology, selecting a high-quality service from a wide range of options has become a complex task. This study aims to address the challenges of data sparsity and the cold-start problem in web service recommendation using Quality of Service (QoS). We propose a novel approach called QoS-aware graph contrastive learning (QAGCL) for web service recommendation. Our model harnesses the power of graph contrastive learning to handle cold-start problems and improve recommendation accuracy effectively. By constructing contextually augmented graphs with geolocation information and randomness, our model provides diverse views. Through the use of graph convolutional networks and graph contrastive learning techniques, we learn user and service embeddings from these augmented graphs. The learned embeddings are then utilized to seamlessly integrate QoS considerations into the recommendation process. Experimental results demonstrate the superiority of our QAGCL model over several existing models, highlighting its effectiveness in addressing data sparsity and the cold-start problem in QoS-aware service recommendations. Our research contributes to the potential for more accurate recommendations in real-world scenarios, even with limited user-service interaction data.

Long-term time series forecasting (LTSF) is a challenging task that has been investigated in various domains such as finance investment, health care, traffic, and weather forecasting. In recent years, Linear-based LTSF models showed better performance, pointing out the problem of Transformer-based approaches causing temporal information loss. However, Linear-based approach has also limitations that the model is too simple to comprehensively exploit the characteristics of the dataset. To solve these limitations, we propose LTSF-DNODE, which applies a model based on linear ordinary differential equations (ODEs) and a time series decomposition method according to data statistical characteristics. We show that LTSF-DNODE outperforms the baselines on various real-world datasets. In addition, for each dataset, we explore the impacts of regularization in the neural ordinary differential equation (NODE) framework.

Graph neural networks (GNNs) are one of the most popular research topics for deep learning. GNN methods typically have been designed on top of the graph signal processing theory. In particular, diffusion equations have been widely used for designing the core processing layer of GNNs, and therefore they are inevitably vulnerable to the notorious oversmoothing problem. Recently, a couple of papers paid attention to reaction equations in conjunctions with diffusion equations. However, they all consider limited forms of reaction equations. To this end, we present a reaction-diffusion equation-based GNN method that considers all popular types of reaction equations in addition to one special reaction equation designed by us. To our knowledge, our paper is one of the most comprehensive studies on reaction-diffusion equation-based GNNs. In our experiments with 9 datasets and 28 baselines, our method, called GREAD, outperforms them in a majority of cases. Further synthetic data experiments show that it mitigates the oversmoothing problem and works well for various homophily rates.

Collaborative filtering is one of the most fundamental topics for recommender systems. Various methods have been proposed for collaborative filtering, ranging from matrix factorization to graph convolutional methods. Being inspired by recent successes of graph filtering-based methods and score-based generative models (SGMs), we present a novel concept of blurring-sharpening process model (BSPM). SGMs and BSPMs share the same processing philosophy that new information can be discovered (e.g., new images are generated in the case of SGMs) while original information is first perturbed and then recovered to its original form. However, SGMs and our BSPMs deal with different types of information, and their optimal perturbation and recovery processes have fundamental discrepancies. Therefore, our BSPMs have different forms from SGMs. In addition, our concept not only theoretically subsumes many existing collaborative filtering models but also outperforms them in terms of Recall and NDCG in the three benchmark datasets, Gowalla, Yelp2018, and Amazon-book. In addition, the processing time of our method is comparable to other fast baselines. Our proposed concept has much potential in the future to be enhanced by designing better blurring (i.e., perturbation) and sharpening (i.e., recovery) processes than what we use in this paper. Our code is available at https://github.com/jeongwhanchoi/BSPM.

Traffic forecasting is one of the most popular spatio-temporal tasks in the field of machine learning. A prevalent approach in the field is to combine graph convolutional networks and recurrent neural networks for the spatio-temporal processing. There has been fierce competition and many novel methods have been proposed. In this article, we present the method of spatio-temporal graph neural rough differential equation (STG-NRDE). Neural rough differential equations (NRDEs) are a breakthrough concept for processing time-series data. Their main concept is to use the log-signature transform to convert a time-series sample into a relatively shorter series of feature vectors. We extend the concept and design two NRDEs: one for the temporal processing and the other for the spatial processing. After that, we combine them into a single framework. We conduct experiments with 6 benchmark datasets and 27 baselines. STG-NRDE shows the best accuracy in all cases, outperforming all those 27 baselines by non-trivial margins.

Traffic forecasting is one of the most popular spatio-temporal tasks in the field of machine learning. A prevalent approach in the field is to combine graph convolutional networks and recurrent neural networks for the spatio-temporal processing. There has been fierce competition and many novel methods have been proposed. In this paper, we present the method of spatio-temporal graph neural controlled differential equation (STG-NCDE). Neural controlled differential equations (NCDEs) are a breakthrough concept for processing sequential data. We extend the concept and design two NCDEs: one for the temporal processing and the other for the spatial processing. After that, we combine them into a single framework. We conduct experiments with 6 benchmark datasets and 20 baselines. STG-NCDE shows the best accuracy in all cases, outperforming all those 20 baselines by non-trivial margins.

Many U.S. metropolitan cities are notorious for their severe shortage of parking spots. To this end, we present a proactive prediction-driven optimization framework to dynamically adjust parking prices. We use state-of-the-art deep learning technologies such as neural ordinary differential equations (NODEs) to design our future parking occupancy rate prediction model given historical occupancy rates and price information. Owing to the continuous and bijective characteristics of NODEs, in addition, we design a one-shot price optimization method given a pre-trained prediction model, which requires only one iteration to find the optimal solution. In other words, we optimize the price input to the pre-trained prediction model to achieve targeted occupancy rates in the parking blocks. We conduct experiments with the data collected in San Francisco and Seattle for years. Our prediction model shows the best accuracy in comparison with various temporal or spatio-temporal forecasting models. Our one-shot optimization method greatly outperforms other black-box and white-box search methods in terms of the search time and always returns the optimal price solution.

There were fierce debates on whether the non-linear embedding propagation of GCNs is appropriate to GCN-based recommender systems. It was recently found that the linear embedding propagation shows better accuracy than the non-linear embedding propagation. Since this phenomenon was discovered especially in recommender systems, it is required that we carefully analyze the linearity and non-linearity issue. In this work, therefore, we revisit the issues of i) which of the linear or non-linear propagation is better and ii) which factors of users/items decide the linearity/non-linearity of the embedding propagation. We propose a novel Hybrid method of linear and non-linear collaborative filtering method (HMLET, pronounced as Hamlet). In our design, there exist both linear and non-linear propagation steps, when processing each user or item node, and our gating module chooses one of them, which results in a hybrid model of the linear and non-linear GCN-based collaborative filtering (CF). The proposed model yields the best accuracy in three public benchmark datasets. Moreover, we classify users/items into the following three classes depending on our gating modules' selections: Full-Non-Linearity (FNL), Partial-Non-Linearity (PNL), and Full-Linearity (FL). We found that there exist strong correlations between nodes' centrality and their class membership, i.e., important user/item nodes exhibit more preferences towards the non-linearity during the propagation steps. To our knowledge, we are the first who design a hybrid method and report the correlation between the graph centrality and the linearity/non-linearity of nodes. All HMLET codes and datasets are available at: https://github.com/qbxlvnf11/HMLET.

Collaborative filtering (CF) is a long-standing problem of recommender systems. Many novel methods have been proposed, ranging from classical matrix factorization to recent graph convolutional network-based approaches. After recent fierce debates, researchers started to focus on linear graph convolutional networks (GCNs) with a layer combination, which show state-of-the-art accuracy in many datasets. In this work, we extend them based on neural ordinary differential equations (NODEs), because the linear GCN concept can be interpreted as a differential equation, and present the method of Learnable-Time ODE-based Collaborative Filtering (LT-OCF). The main novelty in our method is that after redesigning linear GCNs on top of the NODE regime, i) we learn the optimal architecture rather than relying on manually designed ones, ii) we learn smooth ODE solutions that are considered suitable for CF, and iii) we test with various ODE solvers that internally build a diverse set of neural network connections. We also present a novel training method specialized to our method. In our experiments with three benchmark datasets, our method consistently outperforms existing methods in terms of various evaluation metrics. One more important discovery is that our best accuracy was achieved by dense connections.

Owing to the remarkable development of deep learning technology, there have been a series of efforts to build deep learning-based climate models. Whereas most of them utilize recurrent neural networks and/or graph neural networks, we design a novel climate model based on the two concepts, the neural ordinary differential equation (NODE) and the diffusion equation. Many physical processes involving a Brownian motion of particles can be described by the diffusion equation and as a result, it is widely used for modeling climate. On the other hand, neural ordinary differential equations (NODEs) are to learn a latent governing equation of ODE from data. In our presented method, we combine them into a single framework and propose a concept, called neural diffusion equation (NDE). Our NDE, equipped with the diffusion equation and one more additional neural network to model inherent uncertainty, can learn an appropriate latent governing equation that best describes a given climate dataset. In our experiments with two real-world and one synthetic datasets and eleven baselines, our method consistently outperforms existing baselines by non-trivial margins.