Improving Vision Transformers by Overlapping Heads in Multi-Head Self-Attention

Vision Transformers have made remarkable progress in recent years, achieving state-of-the-art performance in most vision tasks. A key component of this success is due to the introduction of the Multi-Head Self-Attention (MHSA) module, which enables each head to learn different representations by applying the attention mechanism independently. In this paper, we empirically demonstrate that Vision Transformers can be further enhanced by overlapping the heads in MHSA. We introduce Multi-Overlapped-Head Self-Attention (MOHSA), where heads are overlapped with their two adjacent heads for queries, keys, and values, while zero-padding is employed for the first and last heads, which have only one neighboring head. Various paradigms for overlapping ratios are proposed to fully investigate the optimal performance of our approach. The proposed approach is evaluated using five Transformer models on four benchmark datasets and yields a significant performance boost. The source code will be made publicly available upon publication.

KA-GNN: Kolmogorov-Arnold Graph Neural Networks for Molecular Property Prediction

Molecular property prediction is a crucial task in the process of Artificial Intelligence-Driven Drug Discovery (AIDD). The challenge of developing models that surpass traditional non-neural network methods continues to be a vibrant area of research. This paper presents a novel graph neural network model-the Kolmogorov-Arnold Network (KAN)-based Graph Neural Network (KA-GNN), which incorporates Fourier series, specifically designed for molecular property prediction. This model maintains the high interpretability characteristic of KAN methods while being extremely efficient in computational resource usage, making it an ideal choice for deployment in resource-constrained environments. Tested and validated on seven public datasets, KA-GNN has shown significant improvements in property predictions over the existing state-of-the-art (SOTA) benchmarks.

Occluded Human Pose Estimation based on Limb Joint Augmentation

Human pose estimation aims at locating the specific joints of humans from the images or videos. While existing deep learning-based methods have achieved high positioning accuracy, they often struggle with generalization in occlusion scenarios. In this paper, we propose an occluded human pose estimation framework based on limb joint augmentation to enhance the generalization ability of the pose estimation model on the occluded human bodies. Specifically, the occlusion blocks are at first employed to randomly cover the limb joints of the human bodies from the training images, imitating the scene where the objects or other people partially occlude the human body. Trained by the augmented samples, the pose estimation model is encouraged to accurately locate the occluded keypoints based on the visible ones. To further enhance the localization ability of the model, this paper constructs a dynamic structure loss function based on limb graphs to explore the distribution of occluded joints by evaluating the dependence between adjacent joints. Extensive experimental evaluations on two occluded datasets, OCHuman and CrowdPose, demonstrate significant performance improvements without additional computation cost during inference.

Robust 3D Point Clouds Classification based on Declarative Defenders

3D point cloud classification requires distinct models from 2D image classification due to the divergent characteristics of the respective input data. While 3D point clouds are unstructured and sparse, 2D images are structured and dense. Bridging the domain gap between these two data types is a non-trivial challenge to enable model interchangeability. Recent research using Lattice Point Classifier (LPC) highlights the feasibility of cross-domain applicability. However, the lattice projection operation in LPC generates 2D images with disconnected projected pixels. In this paper, we explore three distinct algorithms for mapping 3D point clouds into 2D images. Through extensive experiments, we thoroughly examine and analyze their performance and defense mechanisms. Leveraging current large foundation models, we scrutinize the feature disparities between regular 2D images and projected 2D images. The proposed approaches demonstrate superior accuracy and robustness against adversarial attacks. The generative model-based mapping algorithms yield regular 2D images, further minimizing the domain gap from regular 2D classification tasks. The source code is available at https://github.com/KaidongLi/pytorch-LatticePointClassifier.git.

MaskBlur: Spatial and Angular Data Augmentation for Light Field Image Super-Resolution

Data augmentation (DA) is an effective approach for enhancing model performance with limited data, such as light field (LF) image super-resolution (SR). LF images inherently possess rich spatial and angular information. Nonetheless, there is a scarcity of DA methodologies explicitly tailored for LF images, and existing works tend to concentrate solely on either the spatial or angular domain. This paper proposes a novel spatial and angular DA strategy named MaskBlur for LF image SR by concurrently addressing spatial and angular aspects. MaskBlur consists of spatial blur and angular dropout two components. Spatial blur is governed by a spatial mask, which controls where pixels are blurred, i.e., pasting pixels between the low-resolution and high-resolution domains. The angular mask is responsible for angular dropout, i.e., selecting which views to perform the spatial blur operation. By doing so, MaskBlur enables the model to treat pixels differently in the spatial and angular domains when super-resolving LF images rather than blindly treating all pixels equally. Extensive experiments demonstrate the efficacy of MaskBlur in significantly enhancing the performance of existing SR methods. We further extend MaskBlur to other LF image tasks such as denoising, deblurring, low-light enhancement, and real-world SR. Code is publicly available at \url{https://github.com/chaowentao/MaskBlur}.

Depth-Wise Convolutions in Vision Transformers for Efficient Training on Small Datasets

The Vision Transformer (ViT) leverages the Transformer's encoder to capture global information by dividing images into patches and achieves superior performance across various computer vision tasks. However, the self-attention mechanism of ViT captures the global context from the outset, overlooking the inherent relationships between neighboring pixels in images or videos. Transformers mainly focus on global information while ignoring the fine-grained local details. Consequently, ViT lacks inductive bias during image or video dataset training. In contrast, convolutional neural networks (CNNs), with their reliance on local filters, possess an inherent inductive bias, making them more efficient and quicker to converge than ViT with less data. In this paper, we present a lightweight Depth-Wise Convolution module as a shortcut in ViT models, bypassing entire Transformer blocks to ensure the models capture both local and global information with minimal overhead. Additionally, we introduce two architecture variants, allowing the Depth-Wise Convolution modules to be applied to multiple Transformer blocks for parameter savings, and incorporating independent parallel Depth-Wise Convolution modules with different kernels to enhance the acquisition of local information. The proposed approach significantly boosts the performance of ViT models on image classification, object detection and instance segmentation by a large margin, especially on small datasets, as evaluated on CIFAR-10, CIFAR-100, Tiny-ImageNet and ImageNet for image classification, and COCO for object detection and instance segmentation. The source code can be accessed at https://github.com/ZTX-100/Efficient_ViT_with_DW.

Mix-Domain Contrastive Learning for Unpaired H&E-to-IHC Stain Translation

H&E-to-IHC stain translation techniques offer a promising solution for precise cancer diagnosis, especially in low-resource regions where there is a shortage of health professionals and limited access to expensive equipment. Considering the pixel-level misalignment of H&E-IHC image pairs, current research explores the pathological consistency between patches from the same positions of the image pair. However, most of them overemphasize the correspondence between domains or patches, overlooking the side information provided by the non-corresponding objects. In this paper, we propose a Mix-Domain Contrastive Learning (MDCL) method to leverage the supervision information in unpaired H&E-to-IHC stain translation. Specifically, the proposed MDCL method aggregates the inter-domain and intra-domain pathology information by estimating the correlation between the anchor patch and all the patches from the matching images, encouraging the network to learn additional contrastive knowledge from mixed domains. With the mix-domain pathology information aggregation, MDCL enhances the pathological consistency between the corresponding patches and the component discrepancy of the patches from the different positions of the generated IHC image. Extensive experiments on two H&E-to-IHC stain translation datasets, namely MIST and BCI, demonstrate that the proposed method achieves state-of-the-art performance across multiple metrics.

Image-to-Joint Inverse Kinematic of a Supportive Continuum Arm Using Deep Learning

In this work, a deep learning-based technique is used to study the image-to-joint inverse kinematics of a tendon-driven supportive continuum arm. An eye-off-hand configuration is considered by mounting a camera at a fixed pose with respect to the inertial frame attached at the arm base. This camera captures an image for each distinct joint variable at each sampling time to construct the training dataset. This dataset is then employed to adapt a feed-forward deep convolutional neural network, namely the modified VGG-16 model, to estimate the joint variable. One thousand images are recorded to train the deep network, and transfer learning and fine-tuning techniques are applied to the modified VGG-16 to further improve the training. Finally, training is also completed with a larger dataset of images that are affected by various types of noises, changes in illumination, and partial occlusion. The main contribution of this research is the development of an image-to-joint network that can estimate the joint variable given an image of the arm, even if the image is not captured in an ideal condition. The key benefits of this research are twofold: 1) image-to-joint mapping can offer a real-time alternative to computationally complex inverse kinematic mapping through analytical models; and 2) the proposed technique can provide robustness against noise, occlusion, and changes in illumination. The dataset is publicly available on Kaggle.

DEGAP: Dual Event-Guided Adaptive Prefixes for Templated-Based Event Argument Extraction Model with Slot Querying

Recent advancements in event argument extraction (EAE) involve incorporating beneficial auxiliary information into models during training and inference, such as retrieved instances and event templates. Additionally, some studies introduce learnable prefix vectors to models. These methods face three challenges: (1) insufficient utilization of relevant event instances due to deficiencies in retrieval; (2) neglect of important information provided by relevant event templates; (3) the advantages of prefixes are constrained due to their inability to meet the specific informational needs of EAE. In this work, we propose DEGAP, which addresses the above challenges through two simple yet effective components: (1) dual prefixes, where the instance-oriented prefix and template-oriented prefix are trained to learn information from different event instances and templates, respectively, and then provide relevant information as cues to EAE model without retrieval; (2) event-guided adaptive gating mechanism, which guides the prefixes based on the target event to fully leverage their advantages. Extensive experiments demonstrate that our method achieves new state-of-the-art performance on four datasets (ACE05, RAMS, WIKIEVENTS, and MLEE). Further analysis verifies the importance of the proposed design and the effectiveness of the main components.

The intelligent prediction and assessment of financial information risk in the cloud computing model

Cloud computing (cloud computing) is a kind of distributed computing, referring to the network "cloud" will be a huge data calculation and processing program into countless small programs, and then, through the system composed of multiple servers to process and analyze these small programs to get the results and return to the user. This report explores the intersection of cloud computing and financial information processing, identifying risks and challenges faced by financial institutions in adopting cloud technology. It discusses the need for intelligent solutions to enhance data processing efficiency and accuracy while addressing security and privacy concerns. Drawing on regulatory frameworks, the report proposes policy recommendations to mitigate concentration risks associated with cloud computing in the financial industry. By combining intelligent forecasting and evaluation technologies with cloud computing models, the study aims to provide effective solutions for financial data processing and management, facilitating the industry's transition towards digital transformation.