DuSEGO: Dual Second-order Equivariant Graph Ordinary Differential Equation

Graph Neural Networks (GNNs) with equivariant properties have achieved significant success in modeling complex dynamic systems and molecular properties. However, their expressiveness ability is limited by: (1) Existing methods often overlook the over-smoothing issue caused by traditional GNN models, as well as the gradient explosion or vanishing problems in deep GNNs. (2) Most models operate on first-order information, neglecting that the real world often consists of second-order systems, which further limits the model's representation capabilities. To address these issues, we propose the \textbf{Du}al \textbf{S}econd-order \textbf{E}quivariant \textbf{G}raph \textbf{O}rdinary Differential Equation (\method{}) for equivariant representation. Specifically, \method{} apply the dual second-order equivariant graph ordinary differential equations (Graph ODEs) on graph embeddings and node coordinates, simultaneously. Theoretically, we first prove that \method{} maintains the equivariant property. Furthermore, we provide theoretical insights showing that \method{} effectively alleviates the over-smoothing problem in both feature representation and coordinate update. Additionally, we demonstrate that the proposed \method{} mitigates the exploding and vanishing gradients problem, facilitating the training of deep multi-layer GNNs. Extensive experiments on benchmark datasets validate the superiority of the proposed \method{} compared to baselines.

Degree Distribution based Spiking Graph Networks for Domain Adaptation

Spiking Graph Networks (SGNs) have garnered significant attraction from both researchers and industry due to their ability to address energy consumption challenges in graph classification. However, SGNs are only effective for in-distribution data and cannot tackle out-of-distribution data. In this paper, we first propose the domain adaptation problem in SGNs, and introduce a novel framework named Degree-aware Spiking Graph Domain Adaptation for Classification. The proposed DeSGDA addresses the spiking graph domain adaptation problem by three aspects: node degree-aware personalized spiking representation, adversarial feature distribution alignment, and pseudo-label distillation. First, we introduce the personalized spiking representation method for generating degree-dependent spiking signals. Specifically, the threshold of triggering a spike is determined by the node degree, allowing this personalized approach to capture more expressive information for classification. Then, we propose the graph feature distribution alignment module that is adversarially trained using membrane potential against a domain discriminator. Such an alignment module can efficiently maintain high performance and low energy consumption in the case of inconsistent distribution. Additionally, we extract consistent predictions across two spaces to create reliable pseudo-labels, effectively leveraging unlabeled data to enhance graph classification performance. Extensive experiments on benchmark datasets validate the superiority of the proposed DeSGDA compared with competitive baselines.

Improving the Expressiveness of $K$-hop Message-Passing GNNs by Injecting Contextualized Substructure Information

Graph neural networks (GNNs) have become the \textit{de facto} standard for representational learning in graphs, and have achieved state-of-the-art performance in many graph-related tasks; however, it has been shown that the expressive power of standard GNNs are equivalent maximally to 1-dimensional Weisfeiler-Lehman (1-WL) Test. Recently, there is a line of works aiming to enhance the expressive power of graph neural networks. One line of such works aim at developing $K$-hop message-passing GNNs where node representation is updated by aggregating information from not only direct neighbors but all neighbors within $K$-hop of the node. Another line of works leverages subgraph information to enhance the expressive power which is proven to be strictly more powerful than 1-WL test. In this work, we discuss the limitation of $K$-hop message-passing GNNs and propose \textit{substructure encoding function} to uplift the expressive power of any $K$-hop message-passing GNN. We further inject contextualized substructure information to enhance the expressiveness of $K$-hop message-passing GNNs. Our method is provably more powerful than previous works on $K$-hop graph neural networks and 1-WL subgraph GNNs, which is a specific type of subgraph based GNN models, and not less powerful than 3-WL. Empirically, our proposed method set new state-of-the-art performance or achieves comparable performance for a variety of datasets. Our code is available at \url{https://github.com/tianyao-aka/Expresive_K_hop_GNNs}.

Elevating Spectral GNNs through Enhanced Band-pass Filter Approximation

Spectral Graph Neural Networks (GNNs) have attracted great attention due to their capacity to capture patterns in the frequency domains with essential graph filters. Polynomial-based ones (namely poly-GNNs), which approximately construct graph filters with conventional or rational polynomials, are routinely adopted in practice for their substantial performances on graph learning tasks. However, previous poly-GNNs aim at achieving overall lower approximation error on different types of filters, e.g., low-pass and high-pass, but ignore a key question: \textit{which type of filter warrants greater attention for poly-GNNs?} In this paper, we first show that poly-GNN with a better approximation for band-pass graph filters performs better on graph learning tasks. This insight further sheds light on critical issues of existing poly-GNNs, i.e., those poly-GNNs achieve trivial performance in approximating band-pass graph filters, hindering the great potential of poly-GNNs. To tackle the issues, we propose a novel poly-GNN named TrigoNet. TrigoNet constructs different graph filters with novel trigonometric polynomial, and achieves leading performance in approximating band-pass graph filters against other polynomials. By applying Taylor expansion and deserting nonlinearity, TrigoNet achieves noticeable efficiency among baselines. Extensive experiments show the advantages of TrigoNet in both accuracy performances and efficiency.

Weakly Supervised Deep Hyperspherical Quantization for Image Retrieval

Deep quantization methods have shown high efficiency on large-scale image retrieval. However, current models heavily rely on ground-truth information, hindering the application of quantization in label-hungry scenarios. A more realistic demand is to learn from inexhaustible uploaded images that are associated with informal tags provided by amateur users. Though such sketchy tags do not obviously reveal the labels, they actually contain useful semantic information for supervising deep quantization. To this end, we propose Weakly-Supervised Deep Hyperspherical Quantization (WSDHQ), which is the first work to learn deep quantization from weakly tagged images. Specifically, 1) we use word embeddings to represent the tags and enhance their semantic information based on a tag correlation graph. 2) To better preserve semantic information in quantization codes and reduce quantization error, we jointly learn semantics-preserving embeddings and supervised quantizer on hypersphere by employing a well-designed fusion layer and tailor-made loss functions. Extensive experiments show that WSDHQ can achieve state-of-art performance on weakly-supervised compact coding. Code is available at https://github.com/gimpong/AAAI21-WSDHQ.

Spectral GNN via Two-dimensional (2-D) Graph Convolution

Spectral Graph Neural Networks (GNNs) have achieved tremendous success in graph learning. As an essential part of spectral GNNs, spectral graph convolution extracts crucial frequency information in graph data, leading to superior performance of spectral GNNs in downstream tasks. However, in this paper, we show that existing spectral GNNs remain critical drawbacks in performing the spectral graph convolution. Specifically, considering the spectral graph convolution as a construction operation towards target output, we prove that existing popular convolution paradigms cannot construct the target output with mild conditions on input graph signals, causing spectral GNNs to fall into suboptimal solutions. To address the issues, we rethink the spectral graph convolution from a more general two-dimensional (2-D) signal convolution perspective and propose a new convolution paradigm, named 2-D graph convolution. We prove that 2-D graph convolution unifies existing graph convolution paradigms, and is capable to construct arbitrary target output. Based on the proposed 2-D graph convolution, we further propose ChebNet2D, an efficient and effective GNN implementation of 2-D graph convolution through applying Chebyshev interpolation. Extensive experiments on benchmark datasets demonstrate both effectiveness and efficiency of the ChebNet2D.

Contrastive Continual Learning with Importance Sampling and Prototype-Instance Relation Distillation

Recently, because of the high-quality representations of contrastive learning methods, rehearsal-based contrastive continual learning has been proposed to explore how to continually learn transferable representation embeddings to avoid the catastrophic forgetting issue in traditional continual settings. Based on this framework, we propose Contrastive Continual Learning via Importance Sampling (CCLIS) to preserve knowledge by recovering previous data distributions with a new strategy for Replay Buffer Selection (RBS), which minimize estimated variance to save hard negative samples for representation learning with high quality. Furthermore, we present the Prototype-instance Relation Distillation (PRD) loss, a technique designed to maintain the relationship between prototypes and sample representations using a self-distillation process. Experiments on standard continual learning benchmarks reveal that our method notably outperforms existing baselines in terms of knowledge preservation and thereby effectively counteracts catastrophic forgetting in online contexts. The code is available at https://github.com/lijy373/CCLIS.

PASCL: Supervised Contrastive Learning with Perturbative Augmentation for Particle Decay Reconstruction

In high-energy physics, particles produced in collision events decay in a format of a hierarchical tree structure, where only the final decay products can be observed using detectors. However, the large combinatorial space of possible tree structures makes it challenging to recover the actual decay process given a set of final particles. To better analyse the hierarchical tree structure, we propose a graph-based deep learning model to infer the tree structure to reconstruct collision events. In particular, we use a compact matrix representation termed as lowest common ancestor generations (LCAG) matrix, to encode the particle decay tree structure. Then, we introduce a perturbative augmentation technique applied to node features, aiming to mimic experimental uncertainties and increase data diversity. We further propose a supervised graph contrastive learning algorithm to utilize the information of inter-particle relations from multiple decay processes. Extensive experiments show that our proposed supervised graph contrastive learning with perturbative augmentation (PASCL) method outperforms state-of-the-art baseline models on an existing physics-based dataset, significantly improving the reconstruction accuracy. This method provides a more effective training strategy for models with the same parameters and makes way for more accurate and efficient high-energy particle physics data analysis.

CLEX: Continuous Length Extrapolation for Large Language Models

Transformer-based Large Language Models (LLMs) are pioneering advances in many natural language processing tasks, however, their exceptional capabilities are restricted within the preset context window of Transformer. Position Embedding (PE) scaling methods, while effective in extending the context window to a specific length, demonstrate either notable limitations in their extrapolation abilities or sacrificing partial performance within the context window. Length extrapolation methods, although theoretically capable of extending the context window beyond the training sequence length, often underperform in practical long-context applications. To address these challenges, we propose Continuous Length EXtrapolation (CLEX) for LLMs. We generalise the PE scaling approaches to model the continuous dynamics by ordinary differential equations over the length scaling factor, thereby overcoming the constraints of current PE scaling methods designed for specific lengths. Moreover, by extending the dynamics to desired context lengths beyond the training sequence length, CLEX facilitates the length extrapolation with impressive performance in practical tasks. We demonstrate that CLEX can be seamlessly incorporated into LLMs equipped with Rotary Position Embedding, such as LLaMA and GPT-NeoX, with negligible impact on training and inference latency. Experimental results reveal that CLEX can effectively extend the context window to over 4x or almost 8x training length, with no deterioration in performance. Furthermore, when evaluated on the practical LongBench benchmark, our model trained on a 4k length exhibits competitive performance against state-of-the-art open-source models trained on context lengths up to 32k.

Multimodality Representation Learning: A Survey on Evolution, Pretraining and Its Applications

Multimodality Representation Learning, as a technique of learning to embed information from different modalities and their correlations, has achieved remarkable success on a variety of applications, such as Visual Question Answering (VQA), Natural Language for Visual Reasoning (NLVR), and Vision Language Retrieval (VLR). Among these applications, cross-modal interaction and complementary information from different modalities are crucial for advanced models to perform any multimodal task, e.g., understand, recognize, retrieve, or generate optimally. Researchers have proposed diverse methods to address these tasks. The different variants of transformer-based architectures performed extraordinarily on multiple modalities. This survey presents the comprehensive literature on the evolution and enhancement of deep learning multimodal architectures to deal with textual, visual and audio features for diverse cross-modal and modern multimodal tasks. This study summarizes the (i) recent task-specific deep learning methodologies, (ii) the pretraining types and multimodal pretraining objectives, (iii) from state-of-the-art pretrained multimodal approaches to unifying architectures, and (iv) multimodal task categories and possible future improvements that can be devised for better multimodal learning. Moreover, we prepare a dataset section for new researchers that covers most of the benchmarks for pretraining and finetuning. Finally, major challenges, gaps, and potential research topics are explored. A constantly-updated paperlist related to our survey is maintained at https://github.com/marslanm/multimodality-representation-learning.