PyTorchKR Papers

최신 논문

총 3,337편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2107.14222공식 코드 · Papers with Code

Rethinking and Improving Relative Position Encoding for Vision Transformer

Relative position encoding (RPE) is important for transformer to capture sequence ordering of input tokens. General efficacy has been proven in natural language processing. However, in computer vision, its efficacy is not well studied and even remains controversial, e.g., whether relative position encoding can work equally well as absolute position? In order to clarify this, we first review existing relative position encoding methods and analyze their pros and cons when applied in vision transformers. We then propose new relative position encoding methods dedicated to 2D images, called image RPE (iRPE). Our methods consider directional relative distance modeling as well as the interactions between queries and relative position embeddings in self-attention mechanism. The proposed iRPE methods are simple and lightweight. They can be easily plugged into transformer blocks. Experiments demonstrate that solely due to the proposed encoding methods, DeiT and DETR obtain up to 1.5% (top-1 Acc) and 1.3% (mAP) stable improvements over their original versions on ImageNet and COCO respectively, without tuning any extra hyperparameters such as learning rate and weight decay. Our ablation and analysis also yield interesting findings, some of which run counter to previous understanding. Code and models are open-sourced at https://github.com/microsoft/Cream/tree/main/iRPE.

arXiv:2107.13955

Local Structure Matters Most: Perturbation Study in NLU

Recent research analyzing the sensitivity of natural language understanding models to word-order perturbations has shown that neural models are surprisingly insensitive to the order of words. In this paper, we investigate this phenomenon by developing order-altering perturbations on the order of words, subwords, and characters to analyze their effect on neural models' performance on language understanding tasks. We experiment with measuring the impact of perturbations to the local neighborhood of characters and global position of characters in the perturbed texts and observe that perturbation functions found in prior literature only affect the global ordering while the local ordering remains relatively unperturbed. We empirically show that neural models, invariant of their inductive biases, pretraining scheme, or the choice of tokenization, mostly rely on the local structure of text to build understanding and make limited use of the global structure.

arXiv:2107.13327

Ranker-agnostic Contextual Position Bias Estimation

Learning-to-rank (LTR) algorithms are ubiquitous and necessary to explore the extensive catalogs of media providers. To avoid the user examining all the results, its preferences are used to provide a subset of relatively small size. The user preferences can be inferred from the interactions with the presented content if explicit ratings are unavailable. However, directly using implicit feedback can lead to learning wrong relevance models and is known as biased LTR. The mismatch between implicit feedback and true relevances is due to various nuisances, with position bias one of the most relevant. Position bias models consider that the lack of interaction with a presented item is not only attributed to the item being irrelevant but because the item was not examined. This paper introduces a method for modeling the probability of an item being seen in different contexts, e.g., for different users, with a single estimator. Our suggested method, denoted as contextual (EM)-based regression, is ranker-agnostic and able to correctly learn the latent examination probabilities while only using implicit feedback. Our empirical results indicate that the method introduced in this paper outperforms other existing position bias estimators in terms of relative error when the examination probability varies across queries. Moreover, the estimated values provide a ranking performance boost when used to debias the implicit ranking data even if there is no context dependency on the examination probabilities.

arXiv:2107.13067

A Deep Learning Algorithm for Piecewise Linear Interface Construction (PLIC)

Piecewise Linear Interface Construction (PLIC) is frequently used to geometrically reconstruct fluid interfaces in Computational Fluid Dynamics (CFD) modeling of two-phase flows. PLIC reconstructs interfaces from a scalar field that represents the volume fraction of each phase in each computational cell. Given the volume fraction and interface normal, the location of a linear interface is uniquely defined. For a cubic computational cell (3D), the position of the planar interface is determined by intersecting the cube with a plane, such that the volume of the resulting truncated polyhedron cell is equal to the volume fraction. Yet it is geometrically complex to find the exact position of the plane, and it involves calculations that can be a computational bottleneck of many CFD models. However, while the forward problem of 3D PLIC is challenging, the inverse problem, of finding the volume of the truncated polyhedron cell given a defined plane, is simple. In this work, we propose a deep learning model for the solution to the forward problem of PLIC by only making use of its inverse problem. The proposed model is up to several orders of magnitude faster than traditional schemes, which significantly reduces the computational bottleneck of PLIC in CFD simulations.

arXiv:2107.13461

Marine Vehicles Localization Using Grid Cells for Path Integration

Autonomous Underwater Vehicles (AUVs) are platforms used for research and exploration of marine environments. However, these types of vehicles face many challenges that hinder their widespread use in the industry. One of the main limitations is obtaining accurate position estimation, due to the lack of GPS signal underwater. This estimation is usually done with Kalman filters. However, new developments in the neuroscience field have shed light on the mechanisms by which mammals are able to obtain a reliable estimation of their current position based on external and internal motion cues. A new type of neuron, called Grid cells, has been shown to be part of path integration system in the brain. In this article, we show how grid cells can be used for obtaining a position estimation of underwater vehicles. The model of grid cells used requires only the linear velocities together with heading orientation and provides a reliable estimation of the vehicle's position. We provide simulation results for an AUV which show the feasibility of our proposed methodology.

arXiv:2107.12692

Dynamic and Static Object Detection Considering Fusion Regions and Point-wise Features

Object detection is a critical problem for the safe interaction between autonomous vehicles and road users. Deep-learning methodologies allowed the development of object detection approaches with better performance. However, there is still the challenge to obtain more characteristics from the objects detected in real-time. The main reason is that more information from the environment's objects can improve the autonomous vehicle capacity to face different urban situations. This paper proposes a new approach to detect static and dynamic objects in front of an autonomous vehicle. Our approach can also get other characteristics from the objects detected, like their position, velocity, and heading. We develop our proposal fusing results of the environment's interpretations achieved of YoloV3 and a Bayesian filter. To demonstrate our proposal's performance, we asses it through a benchmark dataset and real-world data obtained from an autonomous platform. We compared the results achieved with another approach.

arXiv:2107.12516

They Chose to Not Tell You

The world's stock markets display a strikingly suspicious, decades long pattern of overnight and intraday returns that nobody (other than us) has plausibly explained and that nobody (other than us) has clearly and persistently alerted you to. We use correspondence on this topic over the past five years to show that the silence of others on this issue does not arise from their having a good reason to believe this pattern is fine. Separately, and regardless of whether this pattern turns out to be fine, we have documented that people in a position to alert you to the presence of strikingly suspicious return patterns in the world's stock markets that nobody can innocuously explain are aware of this issue, have no good reason to believe it is not a problem, and chose to not tell you.

arXiv:2107.12800

Deep Reinforcement Learning for L3 Slice Localization in Sarcopenia Assessment

Sarcopenia is a medical condition characterized by a reduction in muscle mass and function. A quantitative diagnosis technique consists of localizing the CT slice passing through the middle of the third lumbar area (L3) and segmenting muscles at this level. In this paper, we propose a deep reinforcement learning method for accurate localization of the L3 CT slice. Our method trains a reinforcement learning agent by incentivizing it to discover the right position. Specifically, a Deep Q-Network is trained to find the best policy to follow for this problem. Visualizing the training process shows that the agent mimics the scrolling of an experienced radiologist. Extensive experiments against other state-of-the-art deep learning based methods for L3 localization prove the superiority of our technique which performs well even with a limited amount of data and annotations.

arXiv:2107.12600

PiSLTRc: Position-informed Sign Language Transformer with Content-aware Convolution

Since the superiority of Transformer in learning long-term dependency, the sign language Transformer model achieves remarkable progress in Sign Language Recognition (SLR) and Translation (SLT). However, there are several issues with the Transformer that prevent it from better sign language understanding. The first issue is that the self-attention mechanism learns sign video representation in a frame-wise manner, neglecting the temporal semantic structure of sign gestures. Secondly, the attention mechanism with absolute position encoding is direction and distance unaware, thus limiting its ability. To address these issues, we propose a new model architecture, namely PiSLTRc, with two distinctive characteristics: (i) content-aware and position-aware convolution layers. Specifically, we explicitly select relevant features using a novel content-aware neighborhood gathering method. Then we aggregate these features with position-informed temporal convolution layers, thus generating robust neighborhood-enhanced sign representation. (ii) injecting the relative position information to the attention mechanism in the encoder, decoder, and even encoder-decoder cross attention. Compared with the vanilla Transformer model, our model performs consistently better on three large-scale sign language benchmarks: PHOENIX-2014, PHOENIX-2014-T and CSL. Furthermore, extensive experiments demonstrate that the proposed method achieves state-of-the-art performance on translation quality with $+1.6$ BLEU improvements.

arXiv:2107.11337

Three-Dimensional Localization of Active Aerial Targets Using a Single Terrestrial Receiver Site

This paper proposes a method for the three-dimensional localization of an active aerial target by a single ground based sensor. The proposed method employs the time and frequency differences of arrival of the signal received directly from the aerial target and the signals received after being reflected from some large auxiliary terrestrial targets (pseudo-sensors) with known positions on the ground. Due to the terrestrial nature of the main and the pseudo sensors, it is impossible to solve for the target's altitude using traditional methods. The proposed method employs target motion analysis to obtain target position including its altitude with acceptable accuracy and low computational complexity. Presented simulations confirm acceptable accuracy of the proposed method in determining three dimensional position of the target despite limited number of the pseudo sensors and its low computational complexity.

arXiv:2107.11185

Where do I rank? Am I happy?: learning income position and subjective-wellbeing in an internet experiment

A tailor-made internet survey experiment provides individuals with information on their income positions to examine their effects on subjective well-being. In the first survey, respondents were asked about their household income and subjective well-being. Based on the data collected, three different respondents' income positions within the residential locality, within a group of the same educational background, and cohort were obtained. In the follow-up survey for the treatment group, respondents are informed of their income positions and then asked for subjective well-being. Key findings are that, after obtaining information, a higher individual's income position improves their subjective well-being. The effects varied according to individual characteristics and proxies.

arXiv:2107.08653공식 코드 · Papers with Code

Cell Detection in Domain Shift Problem Using Pseudo-Cell-Position Heatmap

The domain shift problem is an important issue in automatic cell detection. A detection network trained with training data under a specific condition (source domain) may not work well in data under other conditions (target domain). We propose an unsupervised domain adaptation method for cell detection using the pseudo-cell-position heatmap, where a cell centroid becomes a peak with a Gaussian distribution in the map. In the prediction result for the target domain, even if a peak location is correct, the signal distribution around the peak often has anon-Gaussian shape. The pseudo-cell-position heatmap is re-generated using the peak positions in the predicted heatmap to have a clear Gaussian shape. Our method selects confident pseudo-cell-position heatmaps using a Bayesian network and adds them to the training data in the next iteration. The method can incrementally extend the domain from the source domain to the target domain in a semi-supervised manner. In the experiments using 8 combinations of domains, the proposed method outperformed the existing domain adaptation methods.

arXiv:2107.08067

DeformerNet: A Deep Learning Approach to 3D Deformable Object Manipulation

In this paper, we propose a novel approach to 3D deformable object manipulation leveraging a deep neural network called DeformerNet. Controlling the shape of a 3D object requires an effective state representation that can capture the full 3D geometry of the object. Current methods work around this problem by defining a set of feature points on the object or only deforming the object in 2D image space, which does not truly address the 3D shape control problem. Instead, we explicitly use 3D point clouds as the state representation and apply Convolutional Neural Network on point clouds to learn the 3D features. These features are then mapped to the robot end-effector's position using a fully-connected neural network. Once trained in an end-to-end fashion, DeformerNet directly maps the current point cloud of a deformable object, as well as a target point cloud shape, to the desired displacement in robot gripper position. In addition, we investigate the problem of predicting the manipulation point location given the initial and goal shape of the object.

arXiv:2107.07716

Cooperative Multi-Modal Localization in Connected and Autonomous Vehicles

Cooperative Localization is expected to play a crucial role in various applications in the field of Connected and Autonomous vehicles (CAVs). Future 5G wireless systems are expected to enable cost-effective Vehicle-to-Everything (V2X)systems, allowing CAVs to share with the other entities of the network the data they collect and measure. Typical measurement models usually deployed for this problem, are absolute position from Global Positioning System (GPS), relative distance and azimuth angle to neighbouring vehicles, extracted from Light Detection and Ranging (LIDAR) or Radio Detection and Ranging(RADAR) sensors. In this paper, we provide a cooperative localization approach that performs multi modal-fusion between the interconnected vehicles, by representing a fleet of connected cars as an undirected graph, encoding each vehicle position relative to its neighbouring vehicles. This method is based on:i) the Laplacian Processing, a Graph Signal Processing tool that allows to capture intrinsic geometry of the undirected graph of vehicles rather than their absolute position on global coordinate system and ii) the temporal coherence due to motion patterns of the moving vehicles.

arXiv:2107.05907

Conformer-based End-to-end Speech Recognition With Rotary Position Embedding

Transformer-based end-to-end speech recognition models have received considerable attention in recent years due to their high training speed and ability to model a long-range global context. Position embedding in the transformer architecture is indispensable because it provides supervision for dependency modeling between elements at different positions in the input sequence. To make use of the time order of the input sequence, many works inject some information about the relative or absolute position of the element into the input sequence. In this work, we investigate various position embedding methods in the convolution-augmented transformer (conformer) and adopt a novel implementation named rotary position embedding (RoPE). RoPE encodes absolute positional information into the input sequence by a rotation matrix, and then naturally incorporates explicit relative position information into a self-attention module. To evaluate the effectiveness of the RoPE method, we conducted experiments on AISHELL-1 and LibriSpeech corpora. Results show that the conformer enhanced with RoPE achieves superior performance in the speech recognition task. Specifically, our model achieves a relative word error rate reduction of 8.70% and 7.27% over the conformer on test-clean and test-other sets of the LibriSpeech corpus respectively.

arXiv:2107.05398

Autofocusing Optimal Search Algorithm for a Telescope System

Focus accuracy affects the quality of the astronomical observations. Auto-focusing is necessary for imaging systems designed for astronomical observations. The automatic focus system searches for the best focus position by using a proposed search algorithm. The search algorithm uses the image's focus levels as its objective function in the search process. This paper aims to study the performance of several search algorithms to select a suitable one. The proper search algorithm will be used to develop an automatic focus system for Kottamia Astronomical Observatory (KAO). The optimal search algorithm is selected by applying several search algorithms into five sequences of star-clusters observations. Then, their performance is evaluated based on two criteria, which are accuracy and number of steps. The experimental results show that the Binary search is the optimal search algorithm.

arXiv:2107.05186

Early warning of pedestrians and cyclists

State-of-the-art motor vehicles are able to break for pedestrians in an emergency. We investigate what it would take to issue an early warning to the driver so he/she has time to react. We have identified that predicting the intention of a pedestrian reliably by position is a particularly hard challenge. This paper describes an early pedestrian warning demonstration system.

arXiv:2107.05566공식 코드 · Papers with Code

ProGS: Property Graph Shapes Language (Extended Version)

Property graphs constitute data models for representing knowledge graphs. They allow for the convenient representation of facts, including facts about facts, represented by triples in subject or object position of other triples. Knowledge graphs such as Wikidata are created by a diversity of contributors and a range of sources leaving them prone to two types of errors. The first type of error, falsity of facts, is addressed by property graphs through the representation of provenance and validity, making triples occur as first-order objects in subject position of metadata triples. The second type of error, violation of domain constraints, has not been addressed with regard to property graphs so far. In RDF representations, this error can be addressed by shape languages such as SHACL or ShEx, which allow for checking whether graphs are valid with respect to a set of domain constraints. Borrowing ideas from the syntax and semantics definitions of SHACL, we design a shape language for property graphs, ProGS, which allows for formulating shape constraints on property graphs including their specific constructs, such as edges with identities and key-value annotations to both nodes and edges. We define a formal semantics of ProGS, investigate the resulting complexity of validating property graphs against sets of ProGS shapes, compare with corresponding results for SHACL, and implement a prototypical validator that utilizes answer set programming.

arXiv:2107.05235공식 코드 · Papers with Code

Position-enhanced and Time-aware Graph Convolutional Network for Sequential Recommendations

Most of the existing deep learning-based sequential recommendation approaches utilize the recurrent neural network architecture or self-attention to model the sequential patterns and temporal influence among a user's historical behavior and learn the user's preference at a specific time. However, these methods have two main drawbacks. First, they focus on modeling users' dynamic states from a user-centric perspective and always neglect the dynamics of items over time. Second, most of them deal with only the first-order user-item interactions and do not consider the high-order connectivity between users and items, which has recently been proved helpful for the sequential recommendation. To address the above problems, in this article, we attempt to model user-item interactions by a bipartite graph structure and propose a new recommendation approach based on a Position-enhanced and Time-aware Graph Convolutional Network (PTGCN) for the sequential recommendation. PTGCN models the sequential patterns and temporal dynamics between user-item interactions by defining a position-enhanced and time-aware graph convolution operation and learning the dynamic representations of users and items simultaneously on the bipartite graph with a self-attention aggregator. Also, it realizes the high-order connectivity between users and items by stacking multi-layer graph convolutions. To demonstrate the effectiveness of PTGCN, we carried out a comprehensive evaluation of PTGCN on three real-world datasets of different sizes compared with a few competitive baselines. Experimental results indicate that PTGCN outperforms several state-of-the-art models in terms of two commonly-used evaluation metrics for ranking.

arXiv:2107.04576

Analytical Inverter-Based Distributed Generator Model for Power Flow Analysis

Quantifying the impact of inverter-based distributed generation (DG) sources on power-flow distribution system cases is arduous. Existing distribution system tools predominately model distributed generation sources as either negative PQ loads or as a PV generator and then employed a PV-PQ switching algorithm to mimic Volt/VAR support. These models neglect the unique characteristics of inverter-based distributed generation sources, have scalability and convergence issues, and are ill-suited for increasing solar penetration scenarios. This work proposes an inverter-based DG model accounting for the inverter's topology, sensing position, and control strategies. The model extends recently introduced analytical positive sequence generator models for three-phase studies. The use of circuit-simulation based heuristics help achieve robust convergence. Simulation of the PG&E prototypical feeders using a prototype solver demonstrate the model's accuracy and efficacy.