PyTorchKR Papers

최신 논문

총 139,336편
태그·기간 등 상세 필터

카테고리 탐색

arXiv:2502.18875공식 코드 · Papers with Code

SE(3)-Equivariant Ternary Complex Prediction Towards Target Protein Degradation

Targeted protein degradation (TPD) induced by small molecules has emerged as a rapidly evolving modality in drug discovery, targeting proteins traditionally considered "undruggable". Proteolysis-targeting chimeras (PROTACs) and molecular glue degraders (MGDs) are the primary small molecules that induce TPD. Both types of molecules form a ternary complex linking an E3 ligase with a target protein, a crucial step for drug discovery. While significant advances have been made in binary structure prediction for proteins and small molecules, ternary structure prediction remains challenging due to obscure interaction mechanisms and insufficient training data. Traditional methods relying on manually assigned rules perform poorly and are computationally demanding due to extensive random sampling. In this work, we introduce DeepTernary, a novel deep learning-based approach that directly predicts ternary structures in an end-to-end manner using an encoder-decoder architecture. DeepTernary leverages an SE(3)-equivariant graph neural network (GNN) with both intra-graph and ternary inter-graph attention mechanisms to capture intricate ternary interactions from our collected high-quality training dataset, TernaryDB. The proposed query-based Pocket Points Decoder extracts the 3D structure of the final binding ternary complex from learned ternary embeddings, demonstrating state-of-the-art accuracy and speed in existing PROTAC benchmarks without prior knowledge from known PROTACs. It also achieves notable accuracy on the more challenging MGD benchmark under the blind docking protocol. Remarkably, our experiments reveal that the buried surface area calculated from predicted structures correlates with experimentally obtained degradation potency-related metrics. Consequently, DeepTernary shows potential in effectively assisting and accelerating the development of TPDs for previously undruggable targets.

arXiv:2502.18646공식 코드 · Papers with Code

Spatial Analysis of Neuromuscular Junctions Activation in Three-Dimensional Histology-based Muscle Reconstructions

Histology has long been a foundational technique for studying anatomical structures through tissue slicing. Advances in computational methods now enable three dimensional (3D) reconstruction of organs from histology images, enhancing the analysis of structural and functional features. Here, we present a novel multimodal computational method to reconstruct rodent muscles in 3D using classical image processing and data analysis techniques, analyze their structural features and correlate them to previously recorded electrophysiological data. The algorithm analyzes spatial distribution patterns of features identified through histological staining, normalizing them across multiple samples. Further, the algorithm successfully correlates spatial patterns with high density epimysial ElectroMyoGraphy (hdEMG) recordings, providing a multimodal perspective on neuromuscular dynamics, linking spatial and electrophysiological information. The code was validated by looking at the distribution of NeuroMuscular Junctions (NMJs) in naive soleus muscles and compared the distributions and patterns observed with ones observed in previous literature. Our results showed consistency with the expected results, validating our method for features and pattern recognition. The multimodal aspect was shown in a naive soleus muscle, where a strong correlation was found between motor unit locations derived via hdEMG, and NMJ locations obtained from histology, highlighting their spatial relationship. This multimodal analysis tool integrates 3D structural data with electrophysiological activity, opening new avenues in muscle diagnostics, regenerative medicine, and personalized therapies where spatial insights could one day predict electrophysiological behavior or vice versa.

arXiv:2502.18636공식 코드 · Papers with Code

Transfer Learning Assisted Fast Design Migration Over Technology Nodes: A Study on Transformer Matching Network

In this study, we introduce an innovative methodology for the design of mm-Wave passive networks that leverages knowledge transfer from a pre-trained synthesis neural network (NN) model in one technology node and achieves swift and reliable design adaptation across different integrated circuit (IC) technologies, operating frequencies, and metal options. We prove this concept through simulation-based demonstrations focusing on the training and comparison of the coefficient of determination (R2) of synthesis NNs for 1:1 on-chip transformers in GlobalFoundries(GF) 22nm FDX+ (target domain), with and without transfer learning from a model trained in GF 45nm SOI (source domain). In the experiments, we explore varying target data densities of 0.5%, 1%, 5%, and 100% with a complete dataset of 0.33 million in GF 22FDX+, and for comparative analysis, apply source data densities of 25%, 50%, 75%, and 100% with a complete dataset of 2.5 million in GF 45SOI. With the source data only at 30GHz, the experiments span target data from two metal options in GF 22FDX+ at frequencies of 30 and 39 GHz. The results prove that the transfer learning with the source domain knowledge (GF 45SOI) can both accelerate the training process in the target domain (GF 22FDX+) and improve the R2 values compared to models without knowledge transfer. Furthermore, it is observed that a model trained with just 5% of target data and augmented by transfer learning achieves R2 values superior to a model trained with 20% of the data without transfer, validating the advantage seen from 1% to 5% data density. This demonstrates a notable reduction of 4X in the necessary dataset size highlighting the efficacy of utilizing transfer learning to mm-Wave passive network design. The PyTorch learning and testing code is publicly available at https://github.com/ChenhaoChu/RFIC-TL.

arXiv:2502.20161공식 코드 · Papers with Code

Balanced Rate-Distortion Optimization in Learned Image Compression

Learned image compression (LIC) using deep learning architectures has seen significant advancements, yet standard rate-distortion (R-D) optimization often encounters imbalanced updates due to diverse gradients of the rate and distortion objectives. This imbalance can lead to suboptimal optimization, where one objective dominates, thereby reducing overall compression efficiency. To address this challenge, we reformulate R-D optimization as a multi-objective optimization (MOO) problem and introduce two balanced R-D optimization strategies that adaptively adjust gradient updates to achieve more equitable improvements in both rate and distortion. The first proposed strategy utilizes a coarse-to-fine gradient descent approach along standard R-D optimization trajectories, making it particularly suitable for training LIC models from scratch. The second proposed strategy analytically addresses the reformulated optimization as a quadratic programming problem with an equality constraint, which is ideal for fine-tuning existing models. Experimental results demonstrate that both proposed methods enhance the R-D performance of LIC models, achieving around a 2\% BD-Rate reduction with acceptable additional training cost, leading to a more balanced and efficient optimization process. Code will be available at https://gitlab.com/viper-purdue/Balanced-RD.

arXiv:2502.20100공식 코드 · Papers with Code

Generative augmentations for improved cardiac ultrasound segmentation using diffusion models

One of the main challenges in current research on segmentation in cardiac ultrasound is the lack of large and varied labeled datasets and the differences in annotation conventions between datasets. This makes it difficult to design robust segmentation models that generalize well to external datasets. This work utilizes diffusion models to create generative augmentations that can significantly improve diversity of the dataset and thus the generalisability of segmentation models without the need for more annotated data. The augmentations are applied in addition to regular augmentations. A visual test survey showed that experts cannot clearly distinguish between real and fully generated images. Using the proposed generative augmentations, segmentation robustness was increased when training on an internal dataset and testing on an external dataset with an improvement of over 20 millimeters in Hausdorff distance. Additionally, the limits of agreement for automatic ejection fraction estimation improved by up to 20% of absolute ejection fraction value on out of distribution cases. These improvements come exclusively from the increased variation of the training data using the generative augmentations, without modifying the underlying machine learning model. The augmentation tool is available as an open source Python library at https://github.com/GillesVanDeVyver/EchoGAINS.

arXiv:2502.19949공식 코드 · Papers with Code

Machine-learning for photoplethysmography analysis: Benchmarking feature, image, and signal-based approaches

Photoplethysmography (PPG) is a widely used non-invasive physiological sensing technique, suitable for various clinical applications. Such clinical applications are increasingly supported by machine learning methods, raising the question of the most appropriate input representation and model choice. Comprehensive comparisons, in particular across different input representations, are scarce. We address this gap in the research landscape by a comprehensive benchmarking study covering three kinds of input representations, interpretable features, image representations and raw waveforms, across prototypical regression and classification use cases: blood pressure and atrial fibrillation prediction. In both cases, the best results are achieved by deep neural networks operating on raw time series as input representations. Within this model class, best results are achieved by modern convolutional neural networks (CNNs). but depending on the task setup, shallow CNNs are often also very competitive. We envision that these results will be insightful for researchers to guide their choice on machine learning tasks for PPG data, even beyond the use cases presented in this work.

arXiv:2502.19824공식 코드 · Papers with Code

Shared Stochastic Gaussian Process Latent Variable Models: A Multi-modal Generative Model for Quasar Spectra

This work proposes a scalable probabilistic latent variable model based on Gaussian processes (Lawrence, 2004) in the context of multiple observation spaces. We focus on an application in astrophysics where data sets typically contain both observed spectral features and scientific properties of astrophysical objects such as galaxies or exoplanets. In our application, we study the spectra of very luminous galaxies known as quasars, along with their properties, such as the mass of their central supermassive black hole, accretion rate, and luminosity-resulting in multiple observation spaces. A single data point is then characterized by different classes of observations, each with different likelihoods. Our proposed model extends the baseline stochastic variational Gaussian process latent variable model (GPLVM) introduced by Lalchand et al. (2022) to this setting, proposing a seamless generative model where the quasar spectra and scientific labels can be generated simultaneously using a shared latent space as input to different sets of Gaussian process decoders, one for each observation space. Additionally, this framework enables training in a missing data setting where a large number of dimensions per data point may be unknown or unobserved. We demonstrate high-fidelity reconstructions of the spectra and scientific labels during test-time inference and briefly discuss the scientific interpretations of the results, along with the significance of such a generative model.

arXiv:2502.19659공식 코드 · Papers with Code

Time-Varying Identification of Structural Vector Autoregressions

We propose a novel Bayesian heteroskedastic Markov-switching structural vector autoregression with data-driven time-varying identification. The model selects among alternative patterns of exclusion restrictions to identify structural shocks within the Markov process regimes. We implement the selection through a multinomial prior distribution over these patterns, which is a spike'n'slab prior for individual parameters. By combining a Markov-switching structural matrix with heteroskedastic structural shocks following a stochastic volatility process, the model enables shock identification through time-varying volatility within a regime. As a result, the exclusion restrictions become over-identifying, and their selection is driven by the signal from the data. Our empirical application shows that data support time variation in the US monetary policy shock identification. We also verify that time-varying volatility identifies the monetary policy shock within the regimes.

arXiv:2502.19167공식 코드 · Papers with Code

Generalizable deep learning for photoplethysmography-based blood pressure estimation -- A Benchmarking Study

Photoplethysmography (PPG)-based blood pressure (BP) estimation represents a promising alternative to cuff-based BP measurements. Recently, an increasing number of deep learning models have been proposed to infer BP from the raw PPG waveform. However, these models have been predominantly evaluated on in-distribution test sets, which immediately raises the question of the generalizability of these models to external datasets. To investigate this question, we trained five deep learning models on the recently released PulseDB dataset, provided in-distribution benchmarking results on this dataset, and then assessed out-of-distribution performance on several external datasets. The best model (XResNet1d101) achieved in-distribution MAEs of 9.4 and 6.0 mmHg for systolic and diastolic BP respectively on PulseDB (with subject-specific calibration), and 14.0 and 8.5 mmHg respectively without calibration. Equivalent MAEs on external test datasets without calibration ranged from 15.0 to 25.1 mmHg (SBP) and 7.0 to 10.4 mmHg (DBP). Our results indicate that the performance is strongly influenced by the differences in BP distributions between datasets. We investigated a simple way of improving performance through sample-based domain adaptation and put forward recommendations for training models with good generalization properties. With this work, we hope to educate more researchers for the importance and challenges of out-of-distribution generalization.

arXiv:2502.19081공식 코드 · Papers with Code

Efficient and Accurate Spatial Mixing of Machine Learned Interatomic Potentials for Materials Science

Machine-learned interatomic potentials offer near first-principles accuracy but are computationally expensive, limiting their application in large-scale molecular dynamics simulations. Inspired by quantum mechanics/molecular mechanics methods, we present ML-MIX, an efficient and flexible LAMMPS package for accelerating simulations by spatially mixing interatomic potentials of different complexities. Through constrained linear fitting, we show it is possible to generate a 'cheap' approximate model which closely matches an 'expensive' reference in relevant regions of configuration space. We demonstrate the capability of ML-MIX through case-studies in Si, Fe, and W-He systems, achieving up to an 11x speedup on 8,000 atom systems without sacrificing accuracy on static and dynamic quantities, including calculation of minimum energy paths and dynamical simulations of defect diffusion. For larger domain sizes, we show that the achievable speedup of ML-MIX simulations is limited only by the relative speed of the cheap potential over the expensive potential. The ease of use and flexible nature of this method will extend the practical reach of MLIPs throughout computational materials science, enabling parsimonious application to large spatial and temporal domains.

arXiv:2502.19076공식 코드 · Papers with Code

Circulant ADMM-Net for Fast High-resolution DoA Estimation

This paper introduces CADMM-Net and CHADMM-Net, two deep neural networks for direction of arrival estimation within the least-absolute shrinkage and selection operator (LASSO) framework. These two networks are based on a structured deep unfolding of the alternating direction method of multipliers (ADMM) algorithm through the use of circulant as well as Hermitian-circulant matrices. Along with a computational complexity of $\mathcal{O}(N\log(N))$ per layer for the inference, where $N$ is the length of the dictionary $\mathbf{A}$, they additionally exhibit a memory footprint of $N$ and approximately half of $N$ for CADMMNet and CHADMM-Net, respectively, compared with $N^{2}$ for ADMM-Net. Furthermore, these structured networks exhibit a competitive performance against ADMM-Net, LISTA, TLISTA, and THLISTA with respect to the detection rate, the angular root-mean square error, and the normalized mean squared error.

arXiv:2502.19046공식 코드 · Papers with Code

Max360IQ: Blind Omnidirectional Image Quality Assessment with Multi-axis Attention

Omnidirectional image, also called 360-degree image, is able to capture the entire 360-degree scene, thereby providing more realistic immersive feelings for users than general 2D image and stereoscopic image. Meanwhile, this feature brings great challenges to measuring the perceptual quality of omnidirectional images, which is closely related to users' quality of experience, especially when the omnidirectional images suffer from non-uniform distortion. In this paper, we propose a novel and effective blind omnidirectional image quality assessment (BOIQA) model with multi-axis attention (Max360IQ), which can proficiently measure not only the quality of uniformly distorted omnidirectional images but also the quality of non-uniformly distorted omnidirectional images. Specifically, the proposed Max360IQ is mainly composed of a backbone with stacked multi-axis attention modules for capturing both global and local spatial interactions of extracted viewports, a multi-scale feature integration (MSFI) module to fuse multi-scale features and a quality regression module with deep semantic guidance for predicting the quality of omnidirectional images. Experimental results demonstrate that the proposed Max360IQ outperforms the state-of-the-art Assessor360 by 3.6\% in terms of SRCC on the JUFE database with non-uniform distortion, and gains improvement of 0.4\% and 0.8\% in terms of SRCC on the OIQA and CVIQ databases, respectively. The source code is available at https://github.com/WenJuing/Max360IQ.

arXiv:2502.18998공식 코드 · Papers with Code

Graph Neural Networks embedded into Margules model for vapor-liquid equilibria prediction

Predictive thermodynamic models are crucial for the early stages of product and process design. In this paper the performance of Graph Neural Networks (GNNs) embedded into a relatively simple excess Gibbs energy model, the extended Margules model, for predicting vapor-liquid equilibrium is analyzed. By comparing its performance against the established UNIFAC-Dortmund model it has been shown that GNNs embedded in Margules achieves an overall lower accuracy. However, higher accuracy is observed in the case of various types of binary mixtures. Moreover, since group contribution methods, like UNIFAC, are limited due to feasibility of molecular fragmentation or availability of parameters, the GNN in Margules model offers an alternative for VLE estimation. The findings establish a baseline for the predictive accuracy that simple excess Gibbs energy models combined with GNNs trained solely on infinite dilution data can achieve.

arXiv:2502.18970공식 코드 · Papers with Code

Empirical likelihood approach for high-dimensional moment restrictions with dependent data

Economic and financial models -- such as vector autoregressions, local projections, and multivariate volatility models -- feature complex dynamic interactions and spillovers across many time series. These models can be integrated into a unified framework, with high-dimensional parameters identified by moment conditions. As the number of parameters and moment conditions may surpass the sample size, we propose adding a double penalty to the empirical likelihood criterion to induce sparsity and facilitate dimension reduction. Notably, we utilize a marginal empirical likelihood approach despite temporal dependence in the data. Under regularity conditions, we provide asymptotic guarantees for our method, making it an attractive option for estimating large-scale multivariate time series models. We demonstrate the versatility of our procedure through extensive Monte Carlo simulations and three empirical applications, including analyses of US sectoral inflation rates, fiscal multipliers, and volatility spillover in China's banking sector.

arXiv:2502.18598공식 코드 · Papers with Code

Locational Energy Storage Bid Bounds for Facilitating Social Welfare Convergence

This paper proposes a novel method to generate bid bounds that can serve as offer caps for energy storage in electricity markets to help reduce system costs and regulate potential market power exercises. We derive the bid bounds based on a tractable multi-period economic dispatch chance-constrained formulation that systematically incorporates the uncertainty and risk preference of the system operator. The key analytical results verify that the bounds effectively cap storage bids across all uncertainty scenarios with a guaranteed confidence level. We show that bid bounds decrease as the state of charge increases but rise with greater netload uncertainty and risk preference. We test the effectiveness of the proposed pricing mechanism based on the 8-bus ISO-NE test system, including agent-based storage bidding models. Simulation results demonstrate that the proposed bid bounds effectively align storage bids with the social welfare objective and outperform existing deterministic bid bounds. Under 30% renewable capacity and 20% storage capacity, the bid bounds contribute to an average reduction of 0.17% in system cost, while increasing storage profit by an average of 10.16% across various system uncertainty scenarios and bidding strategies. These benefits scale up with increased storage economic withholding and storage capacity.

arXiv:2502.18575공식 코드 · Papers with Code

Colored Jones Polynomials and the Volume Conjecture

Using the vertex model approach for braid representations, we compute polynomials for spin-1 placed on hyperbolic knots up to 15 crossings. These polynomials are referred to as 3-colored Jones polynomials or adjoint Jones polynomials. Training a subset of the data using a fully connected feedforward neural network, we predict the volume of the knot complement of hyperbolic knots from the adjoint Jones polynomial or its evaluations with 99.34% accuracy. A function of the adjoint Jones polynomial evaluated at the phase $q=e^{ 8 \pi i / 15 }$ predicts the volume with nearly the same accuracy as the neural network. From an analysis of 2-colored and 3-colored Jones polynomials, we conjecture the best phase for $n$-colored Jones polynomials, and use this hypothesis to motivate an improved statement of the volume conjecture. This is tested for knots for which closed form expressions for the $n$-colored Jones polynomial are known, and we show improved convergence to the volume.

arXiv:2502.18533공식 코드 · Papers with Code

Convolutional neural networks for mineral prospecting through alteration mapping with remote sensing data

Traditional geological mapping, based on field observations and rock sample analysis, is inefficient for continuous spatial mapping of features like alteration zones. Deep learning models, such as convolutional neural networks (CNNs), have revolutionised remote sensing data analysis by automatically extracting features for classification and regression tasks. CNNs can detect specific mineralogical changes linked to mineralisation by identifying subtle features in remote sensing data. This study uses CNNs with Landsat 8, Landsat 9, and ASTER data to map alteration zones north of Broken Hill, New South Wales, Australia. The model is trained using ground truth data and an automated approach with selective principal component analysis (PCA). We compare CNNs with traditional machine learning models, including k-nearest neighbours, support vector machines, and multilayer perceptron. Results show that ground truth-based training yields more reliable maps, with CNNs slightly outperforming conventional models in capturing spatial patterns. Landsat 9 outperforms Landsat 8 in mapping iron oxide areas using ground truth-trained CNNs, while ASTER data provides the most accurate argillic and propylitic alteration maps. This highlights CNNs' effectiveness in improving geological mapping precision, especially for identifying subtle mineralisation-related alterations.

arXiv:2502.18523공식 코드 · Papers with Code

End-to-End Deep Learning for Structural Brain Imaging: A Unified Framework

Brain imaging analysis is fundamental in neuroscience, providing valuable insights into brain structure and function. Traditional workflows follow a sequential pipeline-brain extraction, registration, segmentation, parcellation, network generation, and classification-treating each step as an independent task. These methods rely heavily on task-specific training data and expert intervention to correct intermediate errors, making them particularly burdensome for high-dimensional neuroimaging data, where annotations and quality control are costly and time-consuming. We introduce UniBrain, a unified end-to-end framework that integrates all processing steps into a single optimization process, allowing tasks to interact and refine each other. Unlike traditional approaches that require extensive task-specific annotations, UniBrain operates with minimal supervision, leveraging only low-cost labels (i.e., classification and extraction) and a single labeled atlas. By jointly optimizing extraction, registration, segmentation, parcellation, network generation, and classification, UniBrain enhances both accuracy and computational efficiency while significantly reducing annotation effort. Experimental results demonstrate its superiority over existing methods across multiple tasks, offering a more scalable and reliable solution for neuroimaging analysis. Our code and data can be found at https://github.com/Anonymous7852/UniBrain

arXiv:2502.18519공식 코드 · Papers with Code

FreeTumor: Large-Scale Generative Tumor Synthesis in Computed Tomography Images for Improving Tumor Recognition

Tumor is a leading cause of death worldwide, with an estimated 10 million deaths attributed to tumor-related diseases every year. AI-driven tumor recognition unlocks new possibilities for more precise and intelligent tumor screening and diagnosis. However, the progress is heavily hampered by the scarcity of annotated datasets, which demands extensive annotation efforts by radiologists. To tackle this challenge, we introduce FreeTumor, an innovative Generative AI (GAI) framework to enable large-scale tumor synthesis for mitigating data scarcity. Specifically, FreeTumor effectively leverages a combination of limited labeled data and large-scale unlabeled data for tumor synthesis training. Unleashing the power of large-scale data, FreeTumor is capable of synthesizing a large number of realistic tumors on images for augmenting training datasets. To this end, we create the largest training dataset for tumor synthesis and recognition by curating 161,310 publicly available Computed Tomography (CT) volumes from 33 sources, with only 2.3% containing annotated tumors. To validate the fidelity of synthetic tumors, we engaged 13 board-certified radiologists in a Visual Turing Test to discern between synthetic and real tumors. Rigorous clinician evaluation validates the high quality of our synthetic tumors, as they achieved only 51.1% sensitivity and 60.8% accuracy in distinguishing our synthetic tumors from real ones. Through high-quality tumor synthesis, FreeTumor scales up the recognition training datasets by over 40 times, showcasing a notable superiority over state-of-the-art AI methods including various synthesis methods and foundation models. These findings indicate promising prospects of FreeTumor in clinical applications, potentially advancing tumor treatments and improving the survival rates of patients.

arXiv:2502.20392공식 코드 · Papers with Code

Scalable Signature Kernel Computations for Long Time Series via Local Neumann Series Expansions

The signature kernel is a recent state-of-the-art tool for analyzing high-dimensional sequential data, valued for its theoretical guarantees and strong empirical performance. In this paper, we present a novel method for efficiently computing the signature kernel of long, high-dimensional time series via dynamically truncated recursive local power series expansions. Building on the characterization of the signature kernel as the solution of a Goursat PDE, our approach employs tilewise Neumann-series expansions to derive rapidly converging power series approximations of the signature kernel that are locally defined on subdomains and propagated iteratively across the entire domain of the Goursat solution by exploiting the geometry of the time series. Algorithmically, this involves solving a system of interdependent local Goursat PDEs by recursively propagating boundary conditions along a directed graph via topological ordering, with dynamic truncation adaptively terminating each local power series expansion when coefficients fall below machine precision, striking an effective balance between computational cost and accuracy. This method achieves substantial performance improvements over state-of-the-art approaches for computing the signature kernel, providing (a) adjustable and superior accuracy, even for time series with very high roughness; (b) drastically reduced memory requirements; and (c) scalability to efficiently handle very long time series (e.g., with up to half a million points or more) on a single GPU. These advantages make our method particularly well-suited for rough-path-assisted machine learning, financial modeling, and signal processing applications that involve very long and highly volatile data.