PyTorchKR Papers

최신 논문

총 139,336편
태그·기간 등 상세 필터

카테고리 탐색

arXiv:2504.01887공식 코드 · Papers with Code

Dual mechanism of Anti-Seizure Medications in controlling seizure activity

Background: Anti-seizure medications (ASMs) can reduce seizure duration, but their precise modes of action are unclear. Specifically, it is unknown whether ASMs shorten seizures by simply compressing existing seizure activity into a shorter time frame or by selectively suppressing certain seizure activity patterns. Methods: We analysed intracranial EEG (iEEG) recordings of 457 seizures from 28 people with epilepsy undergoing ASM tapering. Beyond measuring seizure occurrence and duration, we categorized distinct seizure activity patterns (states) based on spatial and frequency power characteristics and related these to different ASM levels. Results: We found that reducing ASM levels led to increased seizure frequency (r = 0.87, p < 0.001) and longer seizure duration ($\beta$ = -0.033, p < 0.001), consistent with prior research. Further analysis revealed two distinct mechanisms in which seizures became prolonged: Emergence of new seizure patterns - In approx. 40% of patients, ASM tapering unmasked additional seizure activity states, and seizures containing these 'taper-emergent states' were substantially longer (r = 0.49, p < 0.001). Prolongation of existing seizure patterns - Even in seizures without taper-emergent states, lower ASM levels still resulted in approx. 12-224% longer durations depending on the ASM dosage and tapering ($\beta$ = -0.049, p < 0.001). Conclusion: ASMs influence seizures through two mechanisms: they (i) suppress specific seizure activity patterns (states) in an all-or-nothing fashion and (ii) curtail the duration of other seizure patterns. These findings highlight the complex role of ASMs in seizure modulation and could inform personalized dosing strategies for epilepsy management. These findings may also have implications in understanding the effects of ASMs on cognition and mood.

arXiv:2504.01842공식 코드 · Papers with Code

shapr: Explaining Machine Learning Models with Conditional Shapley Values in R and Python

This paper introduces the shapr R package, a versatile tool for generating Shapley value based prediction explanations for machine learning and statistical regression models. Moreover, the shaprpy Python library brings the core capabilities of shapr to the Python ecosystem. Shapley values originate from cooperative game theory in the 1950s, but have over the past few years become a widely used method for quantifying how a model's features/covariates contribute to specific prediction outcomes. The shapr package emphasizes conditional Shapley value estimates, providing a comprehensive range of approaches for accurately capturing feature dependencies -- a crucial aspect for correct model explanation, typically lacking in similar software. In addition to regular tabular data, the shapr R package includes specialized functionality for explaining time series forecasts. The package offers a minimal set of user functions with sensible default values for most use cases while providing extensive flexibility for advanced users to fine-tune computations. Additional features include parallelized computations, iterative estimation with convergence detection, and rich visualization tools. shapr also extends its functionality to compute causal and asymmetric Shapley values when causal information is available. Overall, the shapr and shaprpy packages aim to enhance the interpretability of predictive models within a powerful and user-friendly framework.

arXiv:2504.01561공식 코드 · Papers with Code

STPNet: Scale-aware Text Prompt Network for Medical Image Segmentation

Accurate segmentation of lesions plays a critical role in medical image analysis and diagnosis. Traditional segmentation approaches that rely solely on visual features often struggle with the inherent uncertainty in lesion distribution and size. To address these issues, we propose STPNet, a Scale-aware Text Prompt Network that leverages vision-language modeling to enhance medical image segmentation. Our approach utilizes multi-scale textual descriptions to guide lesion localization and employs retrieval-segmentation joint learning to bridge the semantic gap between visual and linguistic modalities. Crucially, STPNet retrieves relevant textual information from a specialized medical text repository during training, eliminating the need for text input during inference while retaining the benefits of cross-modal learning. We evaluate STPNet on three datasets: COVID-Xray, COVID-CT, and Kvasir-SEG. Experimental results show that our vision-language approach outperforms state-of-the-art segmentation methods, demonstrating the effectiveness of incorporating textual semantic knowledge into medical image analysis. The code has been made publicly on https://github.com/HUANGLIZI/STPNet.

arXiv:2504.01122공식 코드 · Papers with Code

ffstruc2vec: Flat, Flexible and Scalable Learning of Node Representations from Structural Identities

Node embedding refers to techniques that generate low-dimensional vector representations of nodes in a graph while preserving specific properties of the nodes. A key challenge in the field is developing scalable methods that can preserve structural properties suitable for the required types of structural patterns of a given downstream application task. While most existing methods focus on preserving node proximity, those that do preserve structural properties often lack the flexibility to preserve various types of structural patterns required by downstream application tasks. This paper introduces ffstruc2vec, a scalable deep-learning framework for learning node embedding vectors that preserve structural identities. Its flat, efficient architecture allows high flexibility in capturing diverse types of structural patterns, enabling broad adaptability to various downstream application tasks. The proposed framework significantly outperforms existing approaches across diverse unsupervised and supervised tasks in practical applications. Moreover, ffstruc2vec enables explainability by quantifying how individual structural patterns influence task outcomes, providing actionable interpretation. To our knowledge, no existing framework combines this level of flexibility, scalability, and structural interpretability, underscoring its unique capabilities.

arXiv:2504.00820공식 코드 · Papers with Code

Deep Generative Models: Complexity, Dimensionality, and Approximation

Generative networks have shown remarkable success in learning complex data distributions, particularly in generating high-dimensional data from lower-dimensional inputs. While this capability is well-documented empirically, its theoretical underpinning remains unclear. One common theoretical explanation appeals to the widely accepted manifold hypothesis, which suggests that many real-world datasets, such as images and signals, often possess intrinsic low-dimensional geometric structures. Under this manifold hypothesis, it is widely believed that to approximate a distribution on a $d$-dimensional Riemannian manifold, the latent dimension needs to be at least $d$ or $d+1$. In this work, we show that this requirement on the latent dimension is not necessary by demonstrating that generative networks can approximate distributions on $d$-dimensional Riemannian manifolds from inputs of any arbitrary dimension, even lower than $d$, taking inspiration from the concept of space-filling curves. This approach, in turn, leads to a super-exponential complexity bound of the deep neural networks through expanded neurons. Our findings thus challenge the conventional belief on the relationship between input dimensionality and the ability of generative networks to model data distributions. This novel insight not only corroborates the practical effectiveness of generative networks in handling complex data structures, but also underscores a critical trade-off between approximation error, dimensionality, and model complexity.

arXiv:2504.00599공식 코드 · Papers with Code

Near Field Localization via AI-Aided Subspace Methods

The increasing demands for high-throughput and energy-efficient wireless communications are driving the adoption of extremely large antennas operating at high-frequency bands. In these regimes, multiple users will reside in the radiative near-field, and accurate localization becomes essential. Unlike conventional far-field systems that rely solely on DOA estimation, near-field localization exploits spherical wavefront propagation to recover both DOA and range information. While subspace-based methods, such as MUSIC and its extensions, offer high resolution and interpretability for near-field localization, their performance is significantly impacted by model assumptions, including non-coherent sources, well-calibrated arrays, and a sufficient number of snapshots. To address these limitations, this work proposes AI-aided subspace methods for near-field localization that enhance robustness to real-world challenges. Specifically, we introduce NF-SubspaceNet, a deep learning-augmented 2D MUSIC algorithm that learns a surrogate covariance matrix to improve localization under challenging conditions, and DCD-MUSIC, a cascaded AI-aided approach that decouples angle and range estimation to reduce computational complexity. We further develop a novel model-order-aware training method to accurately estimate the number of sources, that is combined with casting of near field subspace methods as AI models for learning. Extensive simulations demonstrate that the proposed methods outperform classical and existing deep-learning-based localization techniques, providing robust near-field localization even under coherent sources, miscalibrations, and few snapshots.

arXiv:2504.00306공식 코드 · Papers with Code

LOCO-EPI: Leave-one-chromosome-out (LOCO) as a benchmarking paradigm for deep learning based prediction of enhancer-promoter interactions

In mammalian and vertebrate genomes, the promoter regions of the gene and their distal enhancers may be located millions of base-pairs from each other, while a promoter may not interact with the closest enhancer. Since base-pair proximity is not a good indicator of these interactions, there is considerable work toward developing methods for predicting Enhancer-Promoter Interactions (EPI). Several machine learning methods have reported increasingly higher accuracies for predicting EPI. Typically, these approaches randomly split the dataset of Enhancer-Promoter (EP) pairs into training and testing subsets followed by model training. However, the aforementioned random splitting causes information leakage by assigning EP pairs from the same genomic region to both testing and training sets, leading to performance overestimation. In this paper we propose to use a more thorough training and testing paradigm i.e., Leave-one-chromosome-out (LOCO) cross-validation for EPI-prediction. We demonstrate that a deep learning algorithm, which gives higher accuracies when trained and tested on random-splitting setting, drops drastically in performance under LOCO setting, confirming overestimation of performance. We further propose a novel hybrid deep neural network for EPI-prediction that fuses k-mer features of the nucleotide sequence. We show that the hybrid architecture performs significantly better in the LOCO setting, demonstrating it can learn more generalizable aspects of EP interactions. With this paper we are also releasing the LOCO splitting-based EPI dataset. Research data is available in this public repository: https://github.com/malikmtahir/EPI

arXiv:2504.00302공식 코드 · Papers with Code

Deconver: A Deconvolutional Network for Medical Image Segmentation

While convolutional neural networks (CNNs) and vision transformers (ViTs) have advanced medical image segmentation, they face inherent limitations such as local receptive fields in CNNs and high computational complexity in ViTs. This paper introduces Deconver, a novel network that integrates traditional deconvolution techniques from image restoration as a core learnable component within a U-shaped architecture. Deconver replaces computationally expensive attention mechanisms with efficient nonnegative deconvolution (NDC) operations, enabling the restoration of high-frequency details while suppressing artifacts. Key innovations include a backpropagation-friendly NDC layer based on a provably monotonic update rule and a parameter-efficient design. Evaluated across four datasets (ISLES'22, BraTS'23, GlaS, FIVES) covering both 2D and 3D segmentation tasks, Deconver achieves state-of-the-art performance in Dice scores and Hausdorff distance while reducing computational costs (FLOPs) by up to 90% compared to leading baselines. By bridging traditional image restoration with deep learning, this work offers a practical solution for high-precision segmentation in resource-constrained clinical workflows. The project is available at https://github.com/pashtari/deconver.

arXiv:2504.00175공식 코드 · Papers with Code

Exact local recovery for Chemical Shift Imaging

Chemical Shift Imaging (CSI) or Chemical Shift Encoded Magnetic Resonance Imaging (CSE-MRI) enables the quantification of different chemical species in the human body, and it is one of the most widely used imaging modalities used to quantify fat in the human body. Although there have been substantial improvements in the design of signal acquisition protocols and the development of a variety of methods for the recovery of parameters of interest from the measured signal, it is still challenging to obtain a consistent and reliable quantification over the entire field of view. In fact, there are still discrepancies in the quantities recovered by different methods, and each exhibits a different degree of sensitivity to acquisition parameters such as the choice of echo times. Some of these challenges have their origin in the signal model itself. In particular, it is non-linear, and there may be different sets of parameters of interest compatible with the measured signal. For this reason, a thorough analysis of this model may help mitigate some of the remaining challenges, and yield insight into novel acquisition protocols. In this work, we perform an analysis of the signal model underlying CSI, focusing on finding suitable conditions under which recovery of the parameters of interest is possible. We determine the sources of non-identifiability of the parameters, and we propose a reconstruction method based on smooth non-convex optimization under convex constraints that achieves exact local recovery under suitable conditions. A surprising result is that the concentrations of the chemical species in the sample may be identifiable even when other parameters are not. We present numerical results illustrating how our theoretical results may help develop novel acquisition techniques, and showing how our proposed recovery method yields results comparable to the state-of-the-art.

arXiv:2504.00047공식 코드 · Papers with Code

EAP4EMSIG -- Enhancing Event-Driven Microscopy for Microfluidic Single-Cell Analysis

Microfluidic Live-Cell Imaging yields data on microbial cell factories. However, continuous acquisition is challenging as high-throughput experiments often lack realtime insights, delaying responses to stochastic events. We introduce three components in the Experiment Automation Pipeline for Event-Driven Microscopy to Smart Microfluidic Single-Cell Analysis: a fast, accurate Deep Learning autofocusing method predicting the focus offset, an evaluation of real-time segmentation methods and a realtime data analysis dashboard. Our autofocusing achieves a Mean Absolute Error of 0.0226\textmu m with inference times below 50~ms. Among eleven Deep Learning segmentation methods, Cellpose~3 reached a Panoptic Quality of 93.58\%, while a distance-based method is fastest (121~ms, Panoptic Quality 93.02\%). All six Deep Learning Foundation Models were unsuitable for real-time segmentation.

arXiv:2504.00020공식 코드 · Papers with Code

Celler:A Genomic Language Model for Long-Tailed Single-Cell Annotation

Recent breakthroughs in single-cell technology have ushered in unparalleled opportunities to decode the molecular intricacy of intricate biological systems, especially those linked to diseases unique to humans. However, these progressions have also ushered in novel obstacles-specifically, the efficient annotation of extensive, long-tailed single-cell data pertaining to disease conditions. To effectively surmount this challenge, we introduce Celler, a state-of-the-art generative pre-training model crafted specifically for the annotation of single-cell data. Celler incorporates two groundbreaking elements: First, we introduced the Gaussian Inflation (GInf) Loss function. By dynamically adjusting sample weights, GInf Loss significantly enhances the model's ability to learn from rare categories while reducing the risk of overfitting for common categories. Secondly, we introduce an innovative Hard Data Mining (HDM) strategy into the training process, specifically targeting the challenging-to-learn minority data samples, which significantly improved the model's predictive accuracy. Additionally, to further advance research in this field, we have constructed a large-scale single-cell dataset: Celler-75, which encompasses 40 million cells distributed across 80 human tissues and 75 specific diseases. This dataset provides critical support for comprehensively exploring the potential of single-cell technology in disease research. Our code is available at https://github.com/AI4science-ym/HiCeller.

arXiv:2504.01690공식 코드 · Papers with Code

Token Pruning in Audio Transformers: Optimizing Performance and Decoding Patch Importance

Vision Transformers (ViTs) have achieved state-of-the-art performance across various computer vision tasks, but their high computational cost remains a challenge. Token pruning has been proposed to reduce this cost by selectively removing less important tokens. While effective in vision tasks by discarding non-object regions, applying this technique to audio tasks presents unique challenges, as distinguishing relevant from irrelevant regions in time-frequency representations is less straightforward. In this study, for the first time, we applied token pruning to ViT-based audio classification models using Mel-spectrograms and analyzed the trade-offs between model performance and computational cost: TopK token pruning can reduce MAC operations of AudioMAE and AST by 30-40%, with less than a 1% drop in classification accuracy. Our analysis reveals that while high-intensity tokens contribute significantly to model accuracy, low-intensity tokens remain important. In particular, they play a more critical role in general audio classification tasks than in speech-specific tasks.

arXiv:2504.01673공식 코드 · Papers with Code

K-P Quantum Neural Networks

We present an extension of K-P time-optimal quantum control solutions using global Cartan $KAK$ decompositions for geodesic-based solutions. Extending recent time-optimal \emph{constant-$\theta$} control results, we integrate Cartan methods into equivariant quantum neural network (EQNN) for quantum control tasks. We show that a finite-depth limited EQNN ansatz equipped with Cartan layers can replicate the constant-$\theta$ sub-Riemannian geodesics for K-P problems. We demonstrate how for certain classes of control problem on Riemannian symmetric spaces, gradient-based training using an appropriate cost function converges to certain global time-optimal solutions when satisfying simple regularity conditions. This generalises prior geometric control theory methods and clarifies how optimal geodesic estimation can be performed in quantum machine learning contexts.

arXiv:2504.01638공식 코드 · Papers with Code

Convex Computations for Controlled Safety Invariant Sets of Black-box Discrete-time Dynamical Systems

Identifying controlled safety invariant sets (CSISs) is essential in safety-critical applications. This paper tackles the problem of identifying CSISs for black-box discrete-time systems, where the model is unknown and only limited simulation data is accessible. Traditionally, a CSIS is defined as a subset of a safe set, encompassing initial states for which a control input exists that keeps the system within the set at the next time step-this is referred to as the one-step invariance property. However, the requirement for one-step invariance can be equivalently translated into a stricter condition of ``always-invariance'', meaning that there exist control inputs capable of keeping the system within this set indefinitely. Such a condition may prove overly stringent or impractical for black-box systems, where predictions can become unreliable beyond a single time step or a limited number of finite time steps. To overcome the challenges posed by black-box systems, we reformulate the one-step invariance property in a ``Probably Approximately Correct'' (PAC) sense. This approach allows us to assess the probability that a control input exists to keep the system within the CSIS at the next time step, with a predefined level of confidence. If the system successfully remains within the set at the next time step, we can then reapply the invariance evaluation to the new state, thereby facilitating a recursive assurance of invariance. Our method employs barrier functions and scenario optimization, resulting in a linear programming method to estimate PAC CSISs. Finally, the effectiveness of our approach is demonstrated on several examples.

arXiv:2504.01538공식 코드 · Papers with Code

AI-Newton: A Concept-Driven Physical Law Discovery System without Prior Physical Knowledge

Current limitations in human scientific discovery necessitate a new research paradigm. While advances in artificial intelligence (AI) offer a highly promising solution, enabling AI to emulate human-like scientific discovery remains an open challenge. To address this, we propose AI-Newton, a concept-driven discovery system capable of autonomously deriving physical laws from raw data -- without supervision or prior physical knowledge. The system integrates a knowledge base and knowledge representation centered on physical concepts, along with an autonomous discovery workflow. As a proof of concept, we apply AI-Newton to a large set of Newtonian mechanics problems. Given experimental data with noise, the system successfully rediscovers fundamental laws, including Newton's second law, energy conservation and law of gravitation, using autonomously defined concepts. This achievement marks a significant step toward AI-driven autonomous scientific discovery.

arXiv:2504.01532공식 코드 · Papers with Code

Incorporating Coupling Knowledge into Echo State Networks for Learning Spatiotemporally Chaotic Dynamics

Machine learning methods have shown promise in learning chaotic dynamical systems, enabling model-free short-term prediction and attractor reconstruction. However, when applied to large-scale, spatiotemporally chaotic systems, purely data-driven machine learning methods often suffer from inefficiencies, as they require a large learning model size and a massive amount of training data to achieve acceptable performance. To address this challenge, we incorporate the spatial coupling structure of the target system as an inductive bias in the network design. Specifically, we introduce physics-guided clustered echo state networks, leveraging the efficiency of the echo state networks as a base model. Experimental results on benchmark chaotic systems demonstrate that our physics-informed method outperforms existing echo state network models in learning the target chaotic systems. Additionally, our models exhibit robustness to noise in training data and remain effective even when prior coupling knowledge is imperfect. This approach has the potential to enhance other machine learning methods.

arXiv:2504.01250공식 코드 · Papers with Code

R2DN: Scalable Parameterization of Contracting and Lipschitz Recurrent Deep Networks

This paper presents the Robust Recurrent Deep Network (R2DN), a scalable parameterization of robust recurrent neural networks for machine learning and data-driven control. We construct R2DNs as a feedback interconnection of a linear time-invariant system and a 1-Lipschitz deep feedforward network, and directly parameterize the weights so that our models are stable (contracting) and robust to small input perturbations (Lipschitz) by design. Our parameterization uses a structure similar to the previously-proposed recurrent equilibrium networks (RENs), but without the requirement to iteratively solve an equilibrium layer at each time-step. This speeds up model evaluation and backpropagation on GPUs, and makes it computationally feasible to scale up the network size, batch size, and input sequence length in comparison to RENs. We compare R2DNs to RENs on three representative problems in nonlinear system identification, observer design, and learning-based feedback control and find that training and inference are both up to an order of magnitude faster with similar test set performance, and that training/inference times scale more favorably with respect to model expressivity.

arXiv:2504.01093공식 코드 · Papers with Code

Hard-constraining Neumann boundary conditions in physics-informed neural networks via Fourier feature embeddings

We present a novel approach to hard-constrain Neumann boundary conditions in physics-informed neural networks (PINNs) using Fourier feature embeddings. Neumann boundary conditions are used to described critical processes in various application, yet they are more challenging to hard-constrain in PINNs than Dirichlet conditions. Our method employs specific Fourier feature embeddings to directly incorporate Neumann boundary conditions into the neural network's architecture instead of learning them. The embedding can be naturally extended by high frequency modes to better capture high frequency phenomena. We demonstrate the efficacy of our approach through experiments on a diffusion problem, for which our method outperforms existing hard-constraining methods and classical PINNs, particularly in multiscale and high frequency scenarios.

arXiv:2504.01086공식 코드 · Papers with Code

MPCritic: A plug-and-play MPC architecture for reinforcement learning

The reinforcement learning (RL) and model predictive control (MPC) communities have developed vast ecosystems of theoretical approaches and computational tools for solving optimal control problems. Given their conceptual similarities but differing strengths, there has been increasing interest in synergizing RL and MPC. However, existing approaches tend to be limited for various reasons, including computational cost of MPC in an RL algorithm and software hurdles towards seamless integration of MPC and RL tools. These challenges often result in the use of "simple" MPC schemes or RL algorithms, neglecting the state-of-the-art in both areas. This paper presents MPCritic, a machine learning-friendly architecture that interfaces seamlessly with MPC tools. MPCritic utilizes the loss landscape defined by a parameterized MPC problem, focusing on "soft" optimization over batched training steps; thereby updating the MPC parameters while avoiding costly minimization and parametric sensitivities. Since the MPC structure is preserved during training, an MPC agent can be readily used for online deployment, where robust constraint satisfaction is paramount. We demonstrate the versatility of MPCritic, in terms of MPC architectures and RL algorithms that it can accommodate, on classic control benchmarks.

arXiv:2504.01040공식 코드 · Papers with Code

Cal or No Cal? -- Real-Time Miscalibration Detection of LiDAR and Camera Sensors

The goal of extrinsic calibration is the alignment of sensor data to ensure an accurate representation of the surroundings and enable sensor fusion applications. From a safety perspective, sensor calibration is a key enabler of autonomous driving. In the current state of the art, a trend from target-based offline calibration towards targetless online calibration can be observed. However, online calibration is subject to strict real-time and resource constraints which are not met by state-of-the-art methods. This is mainly due to the high number of parameters to estimate, the reliance on geometric features, or the dependence on specific vehicle maneuvers. To meet these requirements and ensure the vehicle's safety at any time, we propose a miscalibration detection framework that shifts the focus from the direct regression of calibration parameters to a binary classification of the calibration state, i.e., calibrated or miscalibrated. Therefore, we propose a contrastive learning approach that compares embedded features in a latent space to classify the calibration state of two different sensor modalities. Moreover, we provide a comprehensive analysis of the feature embeddings and challenging calibration errors that highlight the performance of our approach. As a result, our method outperforms the current state-of-the-art in terms of detection performance, inference time, and resource demand. The code is open source and available on https://github.com/TUMFTM/MiscalibrationDetection.