Understanding the structure of the protein-ligand complex is crucial to drug development. Existing virtual structure measurement and screening methods are dominated by docking and its derived methods combined with deep learning. However, the sampling and scoring methodology have largely restricted the accuracy and efficiency. Here, we show that these two fundamental tasks can be accurately tackled with a single model, namely LigPose, based on multi-task geometric deep learning. By representing the ligand and the protein pair as a graph, LigPose directly optimizes the three-dimensional structure of the complex, with the learning of binding strength and atomic interactions as auxiliary tasks, enabling its one-step prediction ability without docking tools. Extensive experiments show LigPose achieved state-of-the-art performance on major tasks in drug research. Its considerable improvements indicate a promising paradigm of AI-based pipeline for drug development.
We propose an empirical approach centered on the spectral dynamics of weights -- the behavior of singular values and vectors during optimization -- to unify and clarify several phenomena in deep learning. We identify a consistent bias in optimization across various experiments, from small-scale ``grokking'' to large-scale tasks like image classification with ConvNets, image generation with UNets, speech recognition with LSTMs, and language modeling with Transformers. We also demonstrate that weight decay enhances this bias beyond its role as a norm regularizer, even in practical systems. Moreover, we show that these spectral dynamics distinguish memorizing networks from generalizing ones, offering a novel perspective on this longstanding conundrum. Additionally, we leverage spectral dynamics to explore the emergence of well-performing sparse subnetworks (lottery tickets) and the structure of the loss surface through linear mode connectivity. Our findings suggest that spectral dynamics provide a coherent framework to better understand the behavior of neural networks across diverse settings.
In high-energy physics, accurately estimating the kinematic parameters (position and momentum) of anti-neutrons ($\bar{n}$) is essential for exploring the fundamental governing principles. However, this process is particularly challenging when using an electromagnetic calorimeter (EMC) as the energy detector, due to their limited accuracy and efficiency in interacting with $\bar{n}$. To address this issue, we propose Vision Calorimeter (ViC), a data-driven framework which migrates visual object detection techniques to high-energy particle images. To accommodate the unique characteristics of particle images, we introduce the heat-conduction operator (HCO) into both the backbone and the head of the conventional object detector and conduct significant structural improvements. HCO enjoys the advantage of both radial prior and global attention, as it is inspired by physical heat conduction which naturally aligns with the pattern of particle incidence. Implemented via the Discrete Cosine Transform (DCT), HCO extracts frequency-domain features, bridging the distribution gap between the particle images and the natural images on which visual object detectors are pre-trained. Experimental results demonstrate that ViC significantly outperforms traditional approaches, reducing the incident position prediction error by 46.16% (from 17.31$^{\circ}$ to 9.32$^{\circ}$) and providing the first baseline result with an incident momentum regression error of 21.48%. This study underscores ViC's great potential as a general-purpose particle parameter estimator in high-energy physics. Code is available at https://github.com/yuhongtian17/ViC.
Cervical cancer is a crucial global health concern for women, and the persistent infection of High-risk HPV mainly triggers this remains a global health challenge, with young women diagnosis rates soaring from 10\% to 40\% over three decades. While Pap smear screening is a prevalent diagnostic method, visual image analysis can be lengthy and often leads to mistakes. Early detection of the disease can contribute significantly to improving patient outcomes. In recent decades, many researchers have employed machine learning techniques that achieved promise in cervical cancer detection processes based on medical images. In recent years, many researchers have employed various deep-learning techniques to achieve high-performance accuracy in detecting cervical cancer but are still facing various challenges. This research proposes an innovative and novel approach to automate cervical cancer image classification using Multi-Head Self-Attention (MHSA) and convolutional neural networks (CNNs). The proposed method leverages the strengths of both MHSA mechanisms and CNN to effectively capture both local and global features within cervical images in two streams. MHSA facilitates the model's ability to focus on relevant regions of interest, while CNN extracts hierarchical features that contribute to accurate classification. Finally, we combined the two stream features and fed them into the classification module to refine the feature and the classification. To evaluate the performance of the proposed approach, we used the SIPaKMeD dataset, which classifies cervical cells into five categories. Our model achieved a remarkable accuracy of 98.522\%. This performance has high recognition accuracy of medical image classification and holds promise for its applicability in other medical image recognition tasks.
Search and recommendation systems are essential in many services, and they are often developed separately, leading to complex maintenance and technical debt. In this paper, we present a unified deep learning model that efficiently handles key aspects of both tasks.
Concentrating Solar Power (CSP) plants play a crucial role in the global transition towards sustainable energy. A key factor in ensuring the safe and efficient operation of CSP plants is the distribution of concentrated flux density on the receiver. However, the non-ideal flux density generated by individual heliostats can undermine the safety and efficiency of the power plant. The flux density from each heliostat is influenced by its precise surface profile, which includes factors such as canting and mirror errors. Accurately measuring these surface profiles for a large number of heliostats in operation is a formidable challenge. Consequently, control systems often rely on the assumption of ideal surface conditions, which compromises both safety and operational efficiency. In this study, we introduce inverse Deep Learning Ray Tracing (iDLR), an innovative method designed to predict heliostat surfaces based solely on target images obtained during heliostat calibration. Our simulation-based investigation demonstrates that sufficient information regarding the heliostat surface is retained in the flux density distribution of a single heliostat, enabling deep learning models to accurately predict the underlying surface with deflectometry-like precision for the majority of heliostats. Additionally, we assess the limitations of this method, particularly in relation to surface accuracy and resultant flux density predictions. Furthermore, we are presenting a new comprehensive heliostat model using Non-Uniform Rational B-Spline (NURBS) that has the potential to become the new State of the Art for heliostat surface parameterization. Our findings reveal that iDLR has significant potential to enhance CSP plant operations, potentially increasing the overall efficiency and energy output of the power plants.
Hierarchical federated learning (HFL) is a promising distributed deep learning model training paradigm, but it has crucial security concerns arising from adversarial attacks. This research investigates and assesses the security of HFL using a novel methodology by focusing on its resilience against adversarial attacks inference-time and training-time. Through a series of extensive experiments across diverse datasets and attack scenarios, we uncover that HFL demonstrates robustness against untargeted training-time attacks due to its hierarchical structure. However, targeted attacks, particularly backdoor attacks, exploit this architecture, especially when malicious clients are positioned in the overlapping coverage areas of edge servers. Consequently, HFL shows a dual nature in its resilience, showcasing its capability to recover from attacks thanks to its hierarchical aggregation that strengthens its suitability for adversarial training, thereby reinforcing its resistance against inference-time attacks. These insights underscore the necessity for balanced security strategies in HFL systems, leveraging their inherent strengths while effectively mitigating vulnerabilities.
This paper studies the finite sample performance of the flexible estimation approach of Farrell, Liang, and Misra (2021a), who propose to use deep learning for the estimation of heterogeneous parameters in economic models, in the context of discrete choice models. The approach combines the structure imposed by economic models with the flexibility of deep learning, which assures the interpretebility of results on the one hand, and allows estimating flexible functional forms of observed heterogeneity on the other hand. For inference after the estimation with deep learning, Farrell et al. (2021a) derive an influence function that can be applied to many quantities of interest. We conduct a series of Monte Carlo experiments that investigate the impact of regularization on the proposed estimation and inference procedure in the context of discrete choice models. The results show that the deep learning approach generally leads to precise estimates of the true average parameters and that regular robust standard errors lead to invalid inference results, showing the need for the influence function approach for inference. Without regularization, the influence function approach can lead to substantial bias and large estimated standard errors caused by extreme outliers. Regularization reduces this property and stabilizes the estimation procedure, but at the expense of inducing an additional bias. The bias in combination with decreasing variance associated with increasing regularization leads to the construction of invalid inferential statements in our experiments. Repeated sample splitting, unlike regularization, stabilizes the estimation approach without introducing an additional bias, thereby allowing for the construction of valid inferential statements.
Oral squamous cell carcinoma (OSCC) represents a significant global health concern, with increasing incidence rates and challenges in early diagnosis and treatment planning. Early detection is crucial for improving patient outcomes and survival rates. Deep learning, a subset of machine learning, has shown remarkable progress in extracting and analyzing crucial information from medical imaging data.EfficientNetB3, an advanced convolutional neural network architecture, has emerged as a leading model for image classification tasks, including medical imaging. Its superior performance, characterized by high accuracy, precision, and recall, makes it particularly promising for OSCC detection and diagnosis. EfficientNetB3 achieved an accuracy of 0.9833, precision of 0.9782, and recall of 0.9782 in our analysis. By leveraging EfficientNetB3 and other deep learning technologies, clinicians can potentially improve the accuracy and efficiency of OSCC diagnosis, leading to more timely interventions and better patient outcomes. This article also discusses the role of deep learning in advancing precision medicine for OSCC and provides insights into prospects and challenges in leveraging this technology for enhanced cancer care.
In this paper, we investigate a deep learning method for predicting path-dependent processes based on discretely observed historical information. This method is implemented by considering the prediction as a nonparametric regression and obtaining the regression function through simulated samples and deep neural networks. When applying this method to fractional Brownian motion and the solutions of some stochastic differential equations driven by it, we theoretically proved that the $L_2$ errors converge to 0, and we further discussed the scope of the method. With the frequency of discrete observations tending to infinity, the predictions based on discrete observations converge to the predictions based on continuous observations, which implies that we can make approximations by the method. We apply the method to the fractional Brownian motion and the fractional Ornstein-Uhlenbeck process as examples. Comparing the results with the theoretical optimal predictions and taking the mean square error as a measure, the numerical simulations demonstrate that the method can generate accurate results. We also analyze the impact of factors such as prediction period, Hurst index, etc. on the accuracy.
Software development is a repetitive task, as developers usually reuse or get inspiration from existing implementations. Code search, which refers to the retrieval of relevant code snippets from a codebase according to the developer's intent that has been expressed as a query, has become increasingly important in the software development process. Due to the success of deep learning in various applications, a great number of deep learning based code search approaches have sprung up and achieved promising results. However, developers may not follow the same naming conventions and the same variable may have different variable names in different implementations, bringing a challenge to deep learning based code search methods that rely on explicit variable correspondences to understand source code. To overcome this challenge, we propose a naming-agnostic code search method (NACS) based on contrastive multi-view code representation learning. NACS strips information bound to variable names from Abstract Syntax Tree (AST), the representation of the abstract syntactic structure of source code, and focuses on capturing intrinsic properties solely from AST structures. We use semantic-level and syntax-level augmentation techniques to prepare realistically rational data and adopt contrastive learning to design a graph-view modeling component in NACS to enhance the understanding of code snippets. We further model ASTs in a path view to strengthen the graph-view modeling component through multi-view learning. Extensive experiments show that NACS provides superior code search performance compared to baselines and NACS can be adapted to help existing code search methods overcome the impact of different naming conventions.
Research question: We test whether a plane shoulder radiograph can be used together with deep learning methods to identify patients with rotator cuff tears as opposed to using an MRI in standard of care. Findings: By integrating convolutional block attention modules into a deep neural network, our model demonstrates high accuracy in detecting patients with rotator cuff tears, achieving an average AUC of 0.889 and an accuracy of 0.831. Meaning: This study validates the efficacy of our deep learning model to accurately detect rotation cuff tears from radiographs, offering a viable pre-assessment or alternative to more expensive imaging techniques such as MRI.
Deep learning (DL) strategies have recently been utilized to diagnose motor faults by simply analyzing motor phase current signals, offering a less costly and non-intrusive alternative to vibration sensors. This research transforms these time-series current signals into time-frequency 2D representations via Wavelet Transform (WT). The dataset for motor current signals includes 3,750 data points across five categories: one representing normal conditions and four representing artificially induced faults, each under five different load conditions: 0, 25, 50, 75, and 100%. The study employs five WT-based techniques: WT-Amor, WT-Bump, WT-Morse, WSST-Amor, and WSST-Bump. Subsequently, five DL models adopting prior Convolutional Neural Network (CNN) architecture were developed and tested using the transformed 2D plots from each method. The DL models for WT-Amor, WT-Bump, and WT-Morse showed remarkable effectiveness with peak model accuracy of 90.93, 89.20, and 93.73%, respectively, surpassing previous 2D-image-based methods that recorded accuracy of 80.25, 74.80, and 82.80% respectively using the identical dataset and validation protocol. Notably, the WT-Morse approach slightly exceeded the formerly highest ML technique, achieving a 93.20% accuracy. However, the two WSST methods that utilized synchrosqueezing techniques faced difficulty accurately classifying motor faults. The performance of Wavelet-based deep learning methods offers a compelling alternative for machine condition monitoring.
Photovoltaic (PV) energy grows rapidly and is crucial for the decarbonization of electric systems. However, centralized registries recording the technical characteristifs of rooftop PV systems are often missing, making it difficult to accurately monitor this growth. The lack of monitoring could threaten the integration of PV energy into the grid. To avoid this situation, the remote sensing of rooftop PV systems using deep learning emerged as a promising solution. However, existing techniques are not reliable enough to be used by public authorities or transmission system operators (TSOs) to construct up-to-date statistics on the rooftop PV fleet. The lack of reliability comes from the fact that deep learning models are sensitive to distribution shifts. This work proposes a comprehensive evaluation of the effects of distribution shifts on the classification accuracy of deep learning models trained to detect rooftop PV panels on overhead imagery. We construct a benchmark to isolate the sources of distribution shift and introduce a novel methodology that leverages explainable artificial intelligence (XAI) and decomposition of the input image and model's decision in terms of scales to understand how distribution shifts affect deep learning models. Finally, based on our analysis, we introduce a data augmentation technique meant to improve the robustness of deep learning classifiers to varying acquisition conditions. We show that our proposed approach outperforms competing methods. We discuss some practical recommendations for mapping PV systems using overhead imagery and deep learning models.
As global warming soars, the need to assess the environmental impact of research is becoming increasingly urgent. Despite this, few recommender systems research papers address their environmental impact. In this study, we estimate the environmental impact of recommender systems research by reproducing typical experimental pipelines. Our analysis spans 79 full papers from the 2013 and 2023 ACM RecSys conferences, comparing traditional "good old-fashioned AI" algorithms with modern deep learning algorithms. We designed and reproduced representative experimental pipelines for both years, measuring energy consumption with a hardware energy meter and converting it to CO2 equivalents. Our results show that papers using deep learning algorithms emit approximately 42 times more CO2 equivalents than papers using traditional methods. On average, a single deep learning-based paper generates 3,297 kilograms of CO2 equivalents - more than the carbon emissions of one person flying from New York City to Melbourne or the amount of CO2 one tree sequesters over 300 years.
In the modern context, hand gesture recognition has emerged as a focal point. This is due to its wide range of applications, which include comprehending sign language, factories, hands-free devices, and guiding robots. Many researchers have attempted to develop more effective techniques for recognizing these hand gestures. However, there are challenges like dataset limitations, variations in hand forms, external environments, and inconsistent lighting conditions. To address these challenges, we proposed a novel three-stream hybrid model that combines RGB pixel and skeleton-based features to recognize hand gestures. In the procedure, we preprocessed the dataset, including augmentation, to make rotation, translation, and scaling independent systems. We employed a three-stream hybrid model to extract the multi-feature fusion using the power of the deep learning module. In the first stream, we extracted the initial feature using the pre-trained Imagenet module and then enhanced this feature by using a multi-layer of the GRU and LSTM modules. In the second stream, we extracted the initial feature with the pre-trained ReseNet module and enhanced it with the various combinations of the GRU and LSTM modules. In the third stream, we extracted the hand pose key points using the media pipe and then enhanced them using the stacked LSTM to produce the hierarchical feature. After that, we concatenated the three features to produce the final. Finally, we employed a classification module to produce the probabilistic map to generate predicted output. We mainly produced a powerful feature vector by taking advantage of the pixel-based deep learning feature and pos-estimation-based stacked deep learning feature, including a pre-trained model with a scratched deep learning model for unequalled gesture detection capabilities.
Text has become the predominant form of communication on social media, embedding a wealth of emotional nuances. Consequently, the extraction of emotional information from text is of paramount importance. Despite previous research making some progress, existing text sentiment analysis models still face challenges in integrating diverse semantic information and lack interpretability. To address these issues, we propose a quantum-inspired deep learning architecture that combines fundamental principles of quantum mechanics (QM principles) with deep learning models for text sentiment analysis. Specifically, we analyze the commonalities between text representation and QM principles to design a quantum-inspired text representation method and further develop a quantum-inspired text embedding layer. Additionally, we design a feature extraction layer based on long short-term memory (LSTM) networks and self-attention mechanisms (SAMs). Finally, we calculate the text density matrix using the quantum complex numbers principle and apply 2D-convolution neural networks (CNNs) for feature condensation and dimensionality reduction. Through a series of visualization, comparative, and ablation experiments, we demonstrate that our model not only shows significant advantages in accuracy and efficiency compared to previous related models but also achieves a certain level of interpretability by integrating QM principles. Our code is available at QISA.
Deep learning based approaches are now widely used across biophysics to help automate a variety of tasks including image segmentation, feature selection, and deconvolution. However, the presence of multiple competing deep learning architectures, each with its own unique advantages and disadvantages, makes it challenging to select an architecture best suited for a specific application. As such, we present a comprehensive comparison of common models. Here, we focus on the task of segmentation assuming the typically small training dataset sizes available from biophysics experiments and compare the following four commonly used architectures: convolutional neural networks, U-Nets, vision transformers, and vision state space models. In doing so, we establish criteria for determining optimal conditions under which each model excels, thereby offering practical guidelines for researchers and practitioners in the field.
The intersection of artificial intelligence and bioinformatics has enabled significant advancements in drug discovery, particularly through the application of machine learning models. In this study, we present a combined approach using genetic algorithms and deep learning models to address two critical aspects of drug discovery: the generation of novel tyrosine kinase inhibitors and the prediction of their bioactivity. The generative model leverages genetic algorithms to create new small molecules with optimized ADMET (absorption, distribution, metabolism, excretion, and toxicity) and drug-likeness properties. Concurrently, a deep learning model is employed to predict the bioactivity of these generated molecules against tyrosine kinases, a key enzyme family involved in various cellular processes and cancer progression. By integrating these advanced computational methods, we demonstrate a powerful framework for accelerating the generation and identification of potential tyrosine kinase inhibitors, contributing to more efficient and effective early-stage drug discovery processes.
Neural Networks (NN) has been used in many areas with great success. When a NN's structure (Model) is given, during the training steps, the parameters of the model are determined using an appropriate criterion and an optimization algorithm (Training). Then, the trained model can be used for the prediction or inference step (Testing). As there are also many hyperparameters, related to the optimization criteria and optimization algorithms, a validation step is necessary before its final use. One of the great difficulties is the choice of the NN's structure. Even if there are many "on the shelf" networks, selecting or proposing a new appropriate network for a given data, signal or image processing, is still an open problem. In this work, we consider this problem using model based signal and image processing and inverse problems methods. We classify the methods in five classes, based on: i) Explicit analytical solutions, ii) Transform domain decomposition, iii) Operator Decomposition, iv) Optimization algorithms unfolding, and v) Physics Informed NN methods (PINN). Few examples in each category are explained.