PyTorchKR Papers

최신 논문

총 8,922편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2206.04463

Meet You Halfway: Explaining Deep Learning Mysteries

Deep neural networks perform exceptionally well on various learning tasks with state-of-the-art results. While these models are highly expressive and achieve impressively accurate solutions with excellent generalization abilities, they are susceptible to minor perturbations. Samples that suffer such perturbations are known as "adversarial examples". Even though deep learning is an extensively researched field, many questions about the nature of deep learning models remain unanswered. In this paper, we introduce a new conceptual framework attached with a formal description that aims to shed light on the network's behavior and interpret the behind-the-scenes of the learning process. Our framework provides an explanation for inherent questions concerning deep learning. Particularly, we clarify: (1) Why do neural networks acquire generalization abilities? (2) Why do adversarial examples transfer between different models?. We provide a comprehensive set of experiments that support this new framework, as well as its underlying theory.

arXiv:2206.04374공식 코드 · Papers with Code

Uncovering bias in the PlantVillage dataset

We report our investigation on the use of the popular PlantVillage dataset for training deep learning based plant disease detection models. We trained a machine learning model using only 8 pixels from the PlantVillage image backgrounds. The model achieved 49.0% accuracy on the held-out test set, well above the random guessing accuracy of 2.6%. This result indicates that the PlantVillage dataset contains noise correlated with the labels and deep learning models can easily exploit this bias to make predictions. Possible approaches to alleviate this problem are discussed.

arXiv:2206.04394공식 코드 · Papers with Code

Xplique: A Deep Learning Explainability Toolbox

Today's most advanced machine-learning models are hardly scrutable. The key challenge for explainability methods is to help assisting researchers in opening up these black boxes, by revealing the strategy that led to a given decision, by characterizing their internal states or by studying the underlying data representation. To address this challenge, we have developed Xplique: a software library for explainability which includes representative explainability methods as well as associated evaluation metrics. It interfaces with one of the most popular learning libraries: Tensorflow as well as other libraries including PyTorch, scikit-learn and Theano. The code is licensed under the MIT license and is freely available at github.com/deel-ai/xplique.

arXiv:2206.04015

Improving trajectory calculations using deep learning inspired single image superresolution

Lagrangian trajectory or particle dispersion models as well as semi-Lagrangian advection schemes require meteorological data such as wind, temperature and geopotential at the exact spatio-temporal locations of the particles that move independently from a regular grid. Traditionally, this high-resolution data has been obtained by interpolating the meteorological parameters from the gridded data of a meteorological model or reanalysis, e.g. using linear interpolation in space and time. However, interpolation errors are a large source of error for these models. Reducing them requires meteorological input fields with high space and time resolution, which may not always be available and can cause severe data storage and transfer problems. Here, we interpret this problem as a single image superresolution task. We interpret meteorological fields available at their native resolution as low-resolution images and train deep neural networks to up-scale them to higher resolution, thereby providing more accurate data for Lagrangian models. We train various versions of the state-of-the-art Enhanced Deep Residual Networks for Superresolution on low-resolution ERA5 reanalysis data with the goal to up-scale these data to arbitrary spatial resolution. We show that the resulting up-scaled wind fields have root-mean-squared errors half the size of the winds obtained with linear spatial interpolation at acceptable computational inference costs. In a test setup using the Lagrangian particle dispersion model FLEXPART and reduced-resolution wind fields, we demonstrate that absolute horizontal transport deviations of calculated trajectories from "ground-truth" trajectories calculated with undegraded 0.5{\deg} winds are reduced by at least 49.5% (21.8%) after 48 hours relative to trajectories using linear interpolation of the wind data when training on 2{\deg} to 1{\deg} (4{\deg} to 2{\deg}) resolution data.

arXiv:2206.03684

Progress Report: A Deep Learning Guided Exploration of Affine Unimodular Loop Transformations

In this paper, we present a work in progress about a deep learning based approach for automatic code optimization in polyhedral compilers. The proposed technique explores combinations of affine and non-affine loop transformations to find the sequence of transformations that minimizes the execution time of a given program. This exploration is guided by a deep learning based cost model that evaluates the speedup that each sequence of transformations would yield. Preliminary results show that the proposed techniques achieve a 2.35x geometric mean speedup over state of the art polyhedral compilers (Pluto).

arXiv:2206.03900공식 코드 · Papers with Code

Unsupervised Deformable Image Registration with Absent Correspondences in Pre-operative and Post-Recurrence Brain Tumor MRI Scans

Registration of pre-operative and post-recurrence brain images is often needed to evaluate the effectiveness of brain gliomas treatment. While recent deep learning-based deformable registration methods have achieved remarkable success with healthy brain images, most of them would be unable to accurately align images with pathologies due to the absent correspondences in the reference image. In this paper, we propose a deep learning-based deformable registration method that jointly estimates regions with absent correspondence and bidirectional deformation fields. A forward-backward consistency constraint is used to aid in the localization of the resection and recurrence region from voxels with absence correspondences in the two images. Results on 3D clinical data from the BraTS-Reg challenge demonstrate our method can improve image alignment compared to traditional and deep learning-based registration approaches with or without cost function masking strategy. The source code is available at https://github.com/cwmok/DIRAC.

arXiv:2206.03033

Deep Learning Techniques for Visual Counting

In this dissertation, we investigated and enhanced Deep Learning (DL) techniques for counting objects, like pedestrians, cells or vehicles, in still images or video frames. In particular, we tackled the challenge related to the lack of data needed for training current DL-based solutions. Given that the budget for labeling is limited, data scarcity still represents an open problem that prevents the scalability of existing solutions based on the supervised learning of neural networks and that is responsible for a significant drop in performance at inference time when new scenarios are presented to these algorithms. We introduced solutions addressing this issue from several complementary sides, collecting datasets gathered from virtual environments automatically labeled, proposing Domain Adaptation strategies aiming at mitigating the domain gap existing between the training and test data distributions, and presenting a counting strategy in a weakly labeled data scenario, i.e., in the presence of non-negligible disagreement between multiple annotators. Moreover, we tackled the non-trivial engineering challenges coming out of the adoption of Convolutional Neural Network-based techniques in environments with limited power resources, introducing solutions for counting vehicles and pedestrians directly onboard embedded vision systems, i.e., devices equipped with constrained computational capabilities that can capture images and elaborate them.

arXiv:2206.02596

A Robust Deep Learning Enabled Semantic Communication System for Text

With the advent of the 6G era, the concept of semantic communication has attracted increasing attention. Compared with conventional communication systems, semantic communication systems are not only affected by physical noise existing in the wireless communication environment, e.g., additional white Gaussian noise, but also by semantic noise due to the source and the nature of deep learning-based systems. In this paper, we elaborate on the mechanism of semantic noise. In particular, we categorize semantic noise into two categories: literal semantic noise and adversarial semantic noise. The former is caused by written errors or expression ambiguity, while the latter is caused by perturbations or attacks added to the embedding layer via the semantic channel. To prevent semantic noise from influencing semantic communication systems, we present a robust deep learning enabled semantic communication system (R-DeepSC) that leverages a calibrated self-attention mechanism and adversarial training to tackle semantic noise. Compared with baseline models that only consider physical noise for text transmission, the proposed R-DeepSC achieves remarkable performance in dealing with semantic noise under different signal-to-noise ratios.

arXiv:2206.01748공식 코드 · Papers with Code

Federated Deep Learning Meets Autonomous Vehicle Perception: Design and Verification

Realizing human-like perception is a challenge in open driving scenarios due to corner cases and visual occlusions. To gather knowledge of rare and occluded instances, federated learning assisted connected autonomous vehicle (FLCAV) has been proposed, which leverages vehicular networks to establish federated deep neural networks (DNNs) from distributed data captured by vehicles and road sensors. Without the need of data aggregation, FLCAV preserves privacy while reducing communication costs compared with conventional centralized learning. However, it is challenging to determine the network resources and road sensor placements for multi-stage training with multi-modal datasets in multi-variant scenarios. This article presents networking and training frameworks for FLCAV perception. Multi-layer graph resource allocation and vehicle-road contrastive sensor placement are proposed to address the network management and sensor deployment problems, respectively. We also develop CarlaFLCAV, a software platform that implements the above system and methods. Experimental results confirm the superiority of the proposed techniques compared with various benchmarks.

arXiv:2206.02218

Statistical Deep Learning for Spatial and Spatio-Temporal Data

Deep neural network models have become ubiquitous in recent years, and have been applied to nearly all areas of science, engineering, and industry. These models are particularly useful for data that have strong dependencies in space (e.g., images) and time (e.g., sequences). Indeed, deep models have also been extensively used by the statistical community to model spatial and spatio-temporal data through, for example, the use of multi-level Bayesian hierarchical models and deep Gaussian processes. In this review, we first present an overview of traditional statistical and machine learning perspectives for modeling spatial and spatio-temporal data, and then focus on a variety of hybrid models that have recently been developed for latent process, data, and parameter specifications. These hybrid models integrate statistical modeling ideas with deep neural network models in order to take advantage of the strengths of each modeling paradigm. We conclude by giving an overview of computational technologies that have proven useful for these hybrid models, and with a brief discussion on future research directions.

arXiv:2206.02165공식 코드 · Papers with Code

A Survey on Deep Learning based Channel Estimation in Doubly Dispersive Environments

Wireless communications systems are impacted by multi-path fading and Doppler shift in dynamic environments, where the channel becomes doubly-dispersive and its estimation becomes an arduous task. Only a few pilots are used for channel estimation in conventional approaches to preserve high data rate transmission. Consequently, such estimators experience a significant performance degradation in high mobility scenarios. Recently, deep learning has been employed for doubly-dispersive channel estimation due to its low-complexity, robustness, and good generalization ability. Against this backdrop, the current paper presents a comprehensive survey on channel estimation techniques based on deep learning by deeply investigating different methods. The study also provides extensive experimental simulations followed by a computational complexity analysis. After considering different parameters such as modulation order, mobility, frame length, and deep learning architecture, the performance of the studied estimators is evaluated in several mobility scenarios. In addition, the source codes are made available online in order to make the results reproducible.

arXiv:2206.02628

HYCEDIS: HYbrid Confidence Engine for Deep Document Intelligence System

Measuring the confidence of AI models is critical for safely deploying AI in real-world industrial systems. One important application of confidence measurement is information extraction from scanned documents. However, there exists no solution to provide reliable confidence score for current state-of-the-art deep-learning-based information extractors. In this paper, we propose a complete and novel architecture to measure confidence of current deep learning models in document information extraction task. Our architecture consists of a Multi-modal Conformal Predictor and a Variational Cluster-oriented Anomaly Detector, trained to faithfully estimate its confidence on its outputs without the need of host models modification. We evaluate our architecture on real-wold datasets, not only outperforming competing confidence estimators by a huge margin but also demonstrating generalization ability to out-of-distribution data.

arXiv:2206.02342공식 코드 · Papers with Code

WHU-Stereo: A Challenging Benchmark for Stereo Matching of High-Resolution Satellite Images

Stereo matching of high-resolution satellite images (HRSI) is still a fundamental but challenging task in the field of photogrammetry and remote sensing. Recently, deep learning (DL) methods, especially convolutional neural networks (CNNs), have demonstrated tremendous potential for stereo matching on public benchmark datasets. However, datasets for stereo matching of satellite images are scarce. To facilitate further research, this paper creates and publishes a challenging dataset, termed WHU-Stereo, for stereo matching DL network training and testing. This dataset is created by using airborne LiDAR point clouds and high-resolution stereo imageries taken from the Chinese GaoFen-7 satellite (GF-7). The WHU-Stereo dataset contains more than 1700 epipolar rectified image pairs, which cover six areas in China and includes various kinds of landscapes. We have assessed the accuracy of ground-truth disparity maps, and it is proved that our dataset achieves comparable precision compared with existing state-of-the-art stereo matching datasets. To verify its feasibility, in experiments, the hand-crafted SGM stereo matching algorithm and recent deep learning networks have been tested on the WHU-Stereo dataset. Experimental results show that deep learning networks can be well trained and achieves higher performance than hand-crafted SGM algorithm, and the dataset has great potential in remote sensing application. The WHU-Stereo dataset can serve as a challenging benchmark for stereo matching of high-resolution satellite images, and performance evaluation of deep learning models. Our dataset is available at https://github.com/Sheng029/WHU-Stereo

arXiv:2206.01599

A Deep-Learning Usability Expansion Model of Ocean Observations

Today's ocean numerical prediction skills depend on the availability of in-situ and remote ocean observations at the time of the predictions only. Because observations are scarce and discontinuous in time and space, numerical models are often unable to accurately model and predict real ocean dynamics, leading to a lack of fulfillment of a range of services that require reliable predictions at various temporal and spatial scales. The process of constraining free numerical models with observations is known as data assimilation. The primary objective is to minimize the misfit of model states with the observations while respecting the rules of physics. The caveat of this approach is that measurements are used only once, at the time of the prediction. The information contained in the history of the measurements and its role in the determinism of the prediction is, therefore, not accounted for. Consequently, historical measurement cannot be used in real-time forecasting systems. The research presented in this paper provides a novel approach rooted in artificial intelligence to expand the usability of observations made before the time of the prediction. Our approach is based on the re-purpose of an existing deep learning model, called U-Net, designed specifically for image segmentation analysis in the biomedical field. U-Net is used here to create a Transform Model that retains the temporal and spatial evolution of the differences between model and observations to produce a correction in the form of regression weights that evolves spatially and temporally with the model both forward and backward in time, beyond the observation period. Using virtual observations, we show that the usability of the observation can be extended up to a one year prior or post observations.

arXiv:2206.01263

Deep Learning Architecture Based Approach For 2D-Simulation of Microwave Plasma Interaction

This paper presents a convolutional neural network (CNN)-based deep learning model, inspired from UNet with series of encoder and decoder units with skip connections, for the simulation of microwave-plasma interaction. The microwave propagation characteristics in complex plasma medium pertaining to transmission, absorption and reflection primarily depends on the ratio of electromagnetic (EM) wave frequency and electron plasma frequency, and the plasma density profile. The scattering of a plane EM wave with fixed frequency (1 GHz) and amplitude incident on a plasma medium with different gaussian density profiles (in the range of $1\times 10^{17}-1\times 10^{22}{m^{-3}}$) have been considered. The training data associated with microwave-plasma interaction has been generated using 2D-FDTD (Finite Difference Time Domain) based simulations. The trained deep learning model is then used to reproduce the scattered electric field values for the 1GHz incident microwave on different plasma profiles with error margin of less than 2\%. We propose a complete deep learning (DL) based pipeline to train, validate and evaluate the model. We compare the results of the network, using various metrics like SSIM index, average percent error and mean square error, with the physical data obtained from well-established FDTD based EM solvers. To the best of our knowledge, this is the first effort towards exploring a DL based approach for the simulation of complex microwave plasma interaction. The deep learning technique proposed in this work is significantly fast as compared to the existing computational techniques, and can be used as a new, prospective and alternative computational approach for investigating microwave-plasma interaction in a real time scenario.

arXiv:2206.01515

Understanding Deep Learning via Decision Boundary

This paper discovers that the neural network with lower decision boundary (DB) variability has better generalizability. Two new notions, algorithm DB variability and $(\epsilon, \eta)$-data DB variability, are proposed to measure the decision boundary variability from the algorithm and data perspectives. Extensive experiments show significant negative correlations between the decision boundary variability and the generalizability. From the theoretical view, two lower bounds based on algorithm DB variability are proposed and do not explicitly depend on the sample size. We also prove an upper bound of order $\mathcal{O}\left(\frac{1}{\sqrt{m}}+\epsilon+\eta\log\frac{1}{\eta}\right)$ based on data DB variability. The bound is convenient to estimate without the requirement of labels, and does not explicitly depend on the network size which is usually prohibitively large in deep learning.

arXiv:2206.00050공식 코드 · Papers with Code

FiLM-Ensemble: Probabilistic Deep Learning via Feature-wise Linear Modulation

The ability to estimate epistemic uncertainty is often crucial when deploying machine learning in the real world, but modern methods often produce overconfident, uncalibrated uncertainty predictions. A common approach to quantify epistemic uncertainty, usable across a wide class of prediction models, is to train a model ensemble. In a naive implementation, the ensemble approach has high computational cost and high memory demand. This challenges in particular modern deep learning, where even a single deep network is already demanding in terms of compute and memory, and has given rise to a number of attempts to emulate the model ensemble without actually instantiating separate ensemble members. We introduce FiLM-Ensemble, a deep, implicit ensemble method based on the concept of Feature-wise Linear Modulation (FiLM). That technique was originally developed for multi-task learning, with the aim of decoupling different tasks. We show that the idea can be extended to uncertainty quantification: by modulating the network activations of a single deep network with FiLM, one obtains a model ensemble with high diversity, and consequently well-calibrated estimates of epistemic uncertainty, with low computational overhead in comparison. Empirically, FiLM-Ensemble outperforms other implicit ensemble methods, and it and comes very close to the upper bound of an explicit ensemble of networks (sometimes even beating it), at a fraction of the memory cost.

arXiv:2206.00002

Calibrated Bagging Deep Learning for Image Semantic Segmentation: A Case Study on COVID-19 Chest X-ray Image

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) causes coronavirus disease 2019 (COVID-19). Imaging tests such as chest X-ray (CXR) and computed tomography (CT) can provide useful information to clinical staff for facilitating a diagnosis of COVID-19 in a more efficient and comprehensive manner. As a breakthrough of artificial intelligence (AI), deep learning has been applied to perform COVID-19 infection region segmentation and disease classification by analyzing CXR and CT data. However, prediction uncertainty of deep learning models for these tasks, which is very important to safety-critical applications like medical image processing, has not been comprehensively investigated. In this work, we propose a novel ensemble deep learning model through integrating bagging deep learning and model calibration to not only enhance segmentation performance, but also reduce prediction uncertainty. The proposed method has been validated on a large dataset that is associated with CXR image segmentation. Experimental results demonstrate that the proposed method can improve the segmentation performance, as well as decrease prediction uncertainties.

arXiv:2206.00664공식 코드 · Papers with Code

Hopular: Modern Hopfield Networks for Tabular Data

While Deep Learning excels in structured data as encountered in vision and natural language processing, it failed to meet its expectations on tabular data. For tabular data, Support Vector Machines (SVMs), Random Forests, and Gradient Boosting are the best performing techniques with Gradient Boosting in the lead. Recently, we saw a surge of Deep Learning methods that were tailored to tabular data but still underperform compared to Gradient Boosting on small-sized datasets. We suggest "Hopular", a novel Deep Learning architecture for medium- and small-sized datasets, where each layer is equipped with continuous modern Hopfield networks. The modern Hopfield networks use stored data to identify feature-feature, feature-target, and sample-sample dependencies. Hopular's novelty is that every layer can directly access the original input as well as the whole training set via stored data in the Hopfield networks. Therefore, Hopular can step-wise update its current model and the resulting prediction at every layer like standard iterative learning algorithms. In experiments on small-sized tabular datasets with less than 1,000 samples, Hopular surpasses Gradient Boosting, Random Forests, SVMs, and in particular several Deep Learning methods. In experiments on medium-sized tabular data with about 10,000 samples, Hopular outperforms XGBoost, CatBoost, LightGBM and a state-of-the art Deep Learning method designed for tabular data. Thus, Hopular is a strong alternative to these methods on tabular data.