PyTorchKR Papers

최신 논문

총 3,337편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2001.01057

Pixel-Semantic Revise of Position Learning A One-Stage Object Detector with A Shared Encoder-Decoder

Recently, many methods have been proposed for object detection. They cannot detect objects by semantic features, adaptively. In this work, according to channel and spatial attention mechanisms, we mainly analyze that different methods detect objects adaptively. Some state-of-the-art detectors combine different feature pyramids with many mechanisms to enhance multi-level semantic information. However, they require more cost. This work addresses that by an anchor-free detector with shared encoder-decoder with attention mechanism, extracting shared features. We consider features of different levels from backbone (e.g., ResNet-50) as the basis features. Then, we feed the features into a simple module, followed by a detector header to detect objects. Meantime, we use the semantic features to revise geometric locations, and the detector is a pixel-semantic revising of position. More importantly, this work analyzes the impact of different pooling strategies (e.g., mean, maximum or minimum) on multi-scale objects, and finds the minimum pooling improve detection performance on small objects better. Compared with state-of-the-art MNC based on ResNet-101 for the standard MSCOCO 2014 baseline, our method improves detection AP of 3.8%.

arXiv:2001.01618공식 코드 · Papers with Code

ARA : Aggregated RAPPOR and Analysis for Centralized Differential Privacy

Differential privacy(DP) has now become a standard in case of sensitive statistical data analysis. The two main approaches in DP is local and central. Both the approaches have a clear gap in terms of data storing,amount of data to be analyzed, analysis, speed etc. Local wins on the speed. We have tested the state of the art standard RAPPOR which is a local approach and supported this gap. Our work completely focuses on that part too. Here, we propose a model which initially collects RAPPOR reports from multiple clients which are then pushed to a Tf-Idf estimation model. The Tf-Idf estimation model then estimates the reports on the basis of the occurrence of "on bit" in a particular position and its contribution to that position. Thus it generates a centralized differential privacy analysis from multiple clients. Our model successfully and efficiently analyzed the major truth value every time.

arXiv:2001.00841

Glottal Closure and Opening Instant Detection from Speech Signals

This paper proposes a new procedure to detect Glottal Closure and Opening Instants (GCIs and GOIs) directly from speech waveforms. The procedure is divided into two successive steps. First a mean-based signal is computed, and intervals where speech events are expected to occur are extracted from it. Secondly, at each interval a precise position of the speech event is assigned by locating a discontinuity in the Linear Prediction residual. The proposed method is compared to the DYPSA algorithm on the CMU ARCTIC database. A significant improvement as well as a better noise robustness are reported. Besides, results of GOI identification accuracy are promising for the glottal source characterization.

arXiv:2001.00667공식 코드 · Papers with Code

DeepFocus: a Few-Shot Microscope Slide Auto-Focus using a Sample Invariant CNN-based Sharpness Function

Autofocus (AF) methods are extensively used in biomicroscopy, for example to acquire timelapses, where the imaged objects tend to drift out of focus. AD algorithms determine an optimal distance by which to move the sample back into the focal plane. Current hardware-based methods require modifying the microscope and image-based algorithms either rely on many images to converge to the sharpest position or need training data and models specific to each instrument and imaging configuration. Here we propose DeepFocus, an AF method we implemented as a Micro-Manager plugin, and characterize its Convolutional neural network-based sharpness function, which we observed to be depth co-variant and sample-invariant. Sample invariance allows our AF algorithm to converge to an optimal axial position within as few as three iterations using a model trained once for use with a wide range of optical microscopes and a single instrument-dependent calibration stack acquisition of a flat (but arbitrary) textured object. From experiments carried out both on synthetic and experimental data, we observed an average precision, given 3 measured images, of 0.30 +- 0.16 micrometers with a 10x, NA 0.3 objective. We foresee that this performance and low image number will help limit photodamage during acquisitions with light-sensitive samples.

arXiv:2001.00338

Informal Data Transformation Considered Harmful

In this paper we take the common position that AI systems are limited more by the integrity of the data they are learning from than the sophistication of their algorithms, and we take the uncommon position that the solution to achieving better data integrity in the enterprise is not to clean and validate data ex-post-facto whenever needed (the so-called data lake approach to data management, which can lead to data scientists spending 80% of their time cleaning data), but rather to formally and automatically guarantee that data integrity is preserved as it transformed (migrated, integrated, composed, queried, viewed, etc) throughout the enterprise, so that data and programs that depend on that data need not constantly be re-validated for every particular use.

arXiv:2001.00051

Simultaneous Identification of Tweet Purpose and Position

Tweet classification has attracted considerable attention recently. Most of the existing work on tweet classification focuses on topic classification, which classifies tweets into several predefined categories, and sentiment classification, which classifies tweets into positive, negative and neutral. Since tweets are different from conventional text in that they generally are of limited length and contain informal, irregular or new words, so it is difficult to determine user intention to publish a tweet and user attitude towards certain topic. In this paper, we aim to simultaneously classify tweet purpose, i.e., the intention for user to publish a tweet, and position, i.e., supporting, opposing or being neutral to a given topic. By transforming this problem to a multi-label classification problem, a multi-label classification method with post-processing is proposed. Experiments on real-world data sets demonstrate the effectiveness of this method and the results outperform the individual classification methods.

arXiv:1912.12333공식 코드 · Papers with Code

Encoding word order in complex embeddings

Sequential word order is important when processing text. Currently, neural networks (NNs) address this by modeling word position using position embeddings. The problem is that position embeddings capture the position of individual words, but not the ordered relationship (e.g., adjacency or precedence) between individual word positions. We present a novel and principled solution for modeling both the global absolute positions of words and their order relationships. Our solution generalizes word embeddings, previously defined as independent vectors, to continuous word functions over a variable (position). The benefit of continuous functions over variable positions is that word representations shift smoothly with increasing positions. Hence, word representations in different positions can correlate with each other in a continuous function. The general solution of these functions is extended to complex-valued domain due to richer representations. We extend CNN, RNN and Transformer NNs to complex-valued versions to incorporate our complex embedding (we make all code available). Experiments on text classification, machine translation and language modeling show gains over both classical word embeddings and position-enriched word embeddings. To our knowledge, this is the first work in NLP to link imaginary numbers in complex-valued representations to concrete meanings (i.e., word order).

arXiv:1912.12422

TextScanner: Reading Characters in Order for Robust Scene Text Recognition

Driven by deep learning and the large volume of data, scene text recognition has evolved rapidly in recent years. Formerly, RNN-attention based methods have dominated this field, but suffer from the problem of \textit{attention drift} in certain situations. Lately, semantic segmentation based algorithms have proven effective at recognizing text of different forms (horizontal, oriented and curved). However, these methods may produce spurious characters or miss genuine characters, as they rely heavily on a thresholding procedure operated on segmentation maps. To tackle these challenges, we propose in this paper an alternative approach, called TextScanner, for scene text recognition. TextScanner bears three characteristics: (1) Basically, it belongs to the semantic segmentation family, as it generates pixel-wise, multi-channel segmentation maps for character class, position and order; (2) Meanwhile, akin to RNN-attention based methods, it also adopts RNN for context modeling; (3) Moreover, it performs paralleled prediction for character position and class, and ensures that characters are transcripted in correct order. The experiments on standard benchmark datasets demonstrate that TextScanner outperforms the state-of-the-art methods. Moreover, TextScanner shows its superiority in recognizing more difficult text such Chinese transcripts and aligning with target characters.

arXiv:1912.13344공식 코드 · Papers with Code

Learning 3D Human Shape and Pose from Dense Body Parts

Reconstructing 3D human shape and pose from monocular images is challenging despite the promising results achieved by the most recent learning-based methods. The commonly occurred misalignment comes from the facts that the mapping from images to the model space is highly non-linear and the rotation-based pose representation of body models is prone to result in the drift of joint positions. In this work, we investigate learning 3D human shape and pose from dense correspondences of body parts and propose a Decompose-and-aggregate Network (DaNet) to address these issues. DaNet adopts the dense correspondence maps, which densely build a bridge between 2D pixels and 3D vertices, as intermediate representations to facilitate the learning of 2D-to-3D mapping. The prediction modules of DaNet are decomposed into one global stream and multiple local streams to enable global and fine-grained perceptions for the shape and pose predictions, respectively. Messages from local streams are further aggregated to enhance the robust prediction of the rotation-based poses, where a position-aided rotation feature refinement strategy is proposed to exploit spatial relationships between body joints. Moreover, a Part-based Dropout (PartDrop) strategy is introduced to drop out dense information from intermediate representations during training, encouraging the network to focus on more complementary body parts as well as neighboring position features. The efficacy of the proposed method is validated on both indoor and real-world datasets including Human3.6M, UP3D, COCO, and 3DPW, showing that our method could significantly improve the reconstruction performance in comparison with previous state-of-the-art methods. Our code is publicly available at https://hongwenzhang.github.io/dense2mesh .

arXiv:1912.12125

Large-scale 6D Object Pose Estimation Dataset for Industrial Bin-Picking

In this paper, we introduce a new public dataset for 6D object pose estimation and instance segmentation for industrial bin-picking. The dataset comprises both synthetic and real-world scenes. For both, point clouds, depth images, and annotations comprising the 6D pose (position and orientation), a visibility score, and a segmentation mask for each object are provided. Along with the raw data, a method for precisely annotating real-world scenes is proposed. To the best of our knowledge, this is the first public dataset for 6D object pose estimation and instance segmentation for bin-picking containing sufficiently annotated data for learning-based approaches. Furthermore, it is one of the largest public datasets for object pose estimation in general. The dataset is publicly available at http://www.bin-picking.ai/en/dataset.html.

arXiv:1912.11126

An uncertainty principle for neural coding: Conjugate representations of position and velocity are mapped onto firing rates and co-firing rates of neural spike trains

The hippocampal system contains neural populations that encode an animal's position and velocity as it navigates through space. Here, we show that such populations can embed two codes within their spike trains: a firing rate code (R) conveyed by within-cell spike intervals, and a co-firing rate code (R') conveyed by between-cell spike intervals. These two codes behave as conjugates of one another, obeying an analog of the uncertainty principle from physics: information conveyed in R comes at the expense of information in R', and vice versa. An exception to this trade-off occurs when spike trains encode a pair of conjugate variables, such as position and velocity, which do not compete for capacity across R and R'. To illustrate this, we describe two biologically inspired methods for decoding R and R', referred to as sigma and sigma-chi decoding, respectively. Simulations of head direction (HD) and grid cells show that if firing rates are tuned for position (but not velocity), then position is recovered by sigma decoding, whereas velocity is recovered by sigma-chi decoding. Conversely, simulations of oscillatory interference among theta-modulated "speed cells" show that if co-firing rates are tuned for position (but not velocity), then position is recovered by sigma-chi decoding, whereas velocity is recovered by sigma decoding. Between these two extremes, information about both variables can be distributed across both channels, and partially recovered by both decoders. These results suggest that neurons with different spatial and temporal tuning properties-such as speed versus grid cells-might not encode different information, but rather, distribute similar information about position and velocity in different ways across R and R'.

arXiv:1912.10817

Using Prolog for Transforming XML Documents

Proponents of the programming language Prolog share the opinion Prolog is more appropriate for transforming XML-documents as other well-established techniques and languages like XSLT. In order to clarify this position this work proposes a tuProlog-styled interpreter for parsing XML-documents into Prolog-internal lists and vice versa for serialising lists into XML-documents. Based on this implementation a comparison between XSLT and Prolog follows. First, criteria are researched, such as considered language features of XSLT, usability and expressibility. These criteria are validated. Second, it is assessed when Prolog distinguishes between input and output parameters towards reversible transformation.

arXiv:1912.09260

Deep Reinforcement Learning for Motion Planning of Mobile Robots

This paper presents a novel motion and trajectory planning algorithm for nonholonomic mobile robots that uses recent advances in deep reinforcement learning. Starting from a random initial state, i.e., position, velocity and orientation, the robot reaches an arbitrary target state while taking both kinematic and dynamic constraints into account. Our deep reinforcement learning agent not only processes a continuous state space it also executes continuous actions, i.e., the acceleration of wheels and the adaptation of the steering angle. We evaluate our motion and trajectory planning on a mobile robot with a differential drive in a simulation environment.

arXiv:1912.08954

An Adversarial Perturbation Oriented Domain Adaptation Approach for Semantic Segmentation

We focus on Unsupervised Domain Adaptation (UDA) for the task of semantic segmentation. Recently, adversarial alignment has been widely adopted to match the marginal distribution of feature representations across two domains globally. However, this strategy fails in adapting the representations of the tail classes or small objects for semantic segmentation since the alignment objective is dominated by head categories or large objects. In contrast to adversarial alignment, we propose to explicitly train a domain-invariant classifier by generating and defensing against pointwise feature space adversarial perturbations. Specifically, we firstly perturb the intermediate feature maps with several attack objectives (i.e., discriminator and classifier) on each individual position for both domains, and then the classifier is trained to be invariant to the perturbations. By perturbing each position individually, our model treats each location evenly regardless of the category or object size and thus circumvents the aforementioned issue. Moreover, the domain gap in feature space is reduced by extrapolating source and target perturbed features towards each other with attack on the domain discriminator. Our approach achieves the state-of-the-art performance on two challenging domain adaptation tasks for semantic segmentation: GTA5 -> Cityscapes and SYNTHIA -> Cityscapes.

arXiv:1912.08740

Effective Integration of Symbolic and Connectionist Approaches through a Hybrid Representation

In this paper, we present our position for a neuralsymbolic integration strategy, arguing in favor of a hybrid representation to promote an effective integration. Such description differs from others fundamentally, since its entities aim at representing AI models in general, allowing to describe both nonsymbolic and symbolic knowledge, the integration between them and their corresponding processors. Moreover, the entities also support representing workflows, leveraging traceability to keep track of every change applied to models and their related entities (e.g., data or concepts) throughout the lifecycle of the models.

arXiv:1912.08282공식 코드 · Papers with Code

Chinese Named Entity Recognition Augmented with Lexicon Memory

Inspired by a concept of content-addressable retrieval from cognitive science, we propose a novel fragment-based model augmented with a lexicon-based memory for Chinese NER, in which both the character-level and word-level features are combined to generate better feature representations for possible name candidates. It is observed that locating the boundary information of entity names is useful in order to classify them into pre-defined categories. Position-dependent features, including prefix and suffix are introduced for NER in the form of distributed representation. The lexicon-based memory is used to help generate such position-dependent features and deal with the problem of out-of-vocabulary words. Experimental results showed that the proposed model, called LEMON, achieved state-of-the-art on four datasets.

arXiv:1912.07224공식 코드 · Papers with Code

Domain Knowledge Based Brain Tumor Segmentation and Overall Survival Prediction

Automatically segmenting sub-regions of gliomas (necrosis, edema and enhancing tumor) and accurately predicting overall survival (OS) time from multimodal MRI sequences have important clinical significance in diagnosis, prognosis and treatment of gliomas. However, due to the high degree variations of heterogeneous appearance and individual physical state, the segmentation of sub-regions and OS prediction are very challenging. To deal with these challenges, we utilize a 3D dilated multi-fiber network (DMFNet) with weighted dice loss for brain tumor segmentation, which incorporates prior volume statistic knowledge and obtains a balance between small and large objects in MRI scans. For OS prediction, we propose a DenseNet based 3D neural network with position encoding convolutional layer (PECL) to extract meaningful features from T1 contrast MRI, T2 MRI and previously segmented subregions. Both labeled data and unlabeled data are utilized to prevent over-fitting for semi-supervised learning. Those learned deep features along with handcrafted features (such as ages, volume of tumor) and position encoding segmentation features are fed to a Gradient Boosting Decision Tree (GBDT) to predict a specific OS day

arXiv:1912.06365

Fast Image Caption Generation with Position Alignment

Recent neural network models for image captioning usually employ an encoder-decoder architecture, where the decoder adopts a recursive sequence decoding way. However, such autoregressive decoding may result in sequential error accumulation and slow generation which limit the applications in practice. Non-autoregressive (NA) decoding has been proposed to cover these issues but suffers from language quality problem due to the indirect modeling of the target distribution. Towards that end, we propose an improved NA prediction framework to accelerate image captioning. Our decoding part consists of a position alignment to order the words that describe the content detected in the given image, and a fine non-autoregressive decoder to generate elegant descriptions. Furthermore, we introduce an inference strategy that regards position information as a latent variable to guide the further sentence generation. The Experimental results on public datasets show that our proposed model achieves better performance compared to general NA captioning models, while achieves comparable performance as autoregressive image captioning models with a significant speedup.

arXiv:1912.04672

Effects of lead position, cardiac rhythm variation and drug-induced QT prolongation on performance of machine learning methods for ECG processing

Machine learning shows great performance in various problems of electrocardiography (ECG) signal analysis. However, collecting a dataset for biomedical engineering is a very difficult task. Any dataset for ECG processing contains from 100 to 10,000 times fewer cases than datasets for image or text analysis. This issue is especially important because of physiological phenomena that can significantly change the morphology of heartbeats in ECG signals. In this preliminary study, we analyze the effects of lead choice from the standard ECG recordings, variation of ECG during 24-hours, and the effects of QT-prolongation agents on the performance of machine learning methods for ECG processing. We choose the problem of subject identification for analysis, because this problem may be solved for almost any available dataset of ECG data. In a discussion, we compare our findings with observations from other works that use machine learning for ECG processing with different problem statements. Our results show the importance of training dataset enrichment with ECG signals acquired in specific physiological conditions for obtaining good performance of ECG processing for real applications.