arXiv subjects

Jiwon Seo

Publications and source records attributed to Jiwon Seo.

At least 37 records · Page 2Linked to original sources

arXiv2024

Efficient Quantum Circuit Encoding of Object Information in 2D Ray Casting

Quantum computing holds the potential to solve problems that are practically unsolvable by classical computers due to its ability to significantly reduce time complexity. We aim to harness this potential to enhance ray casting, a pivotal technique in computer graphics for simplifying the rendering of 3D objects. To perform ray casting in a quantum computer, we need to encode the defining parameters of primitives into qubits. However, during the current noisy intermediate-scale quantum (NISQ) era, challenges arise from the limited number of qubits and the impact of noise when executing multiple gates. Through logic optimization, we reduced the depth of quantum circuits as well as the number of gates and qubits. As a result, the event count of correct measurements from an IBM quantum computer significantly exceeded that of incorrect measurements.

quant-ph↗

arXiv2024

ExeGPT: Constraint-Aware Resource Scheduling for LLM Inference

This paper presents ExeGPT, a distributed system designed for constraint-aware LLM inference. ExeGPT finds and runs with an optimal execution schedule to maximize inference throughput while satisfying a given latency constraint. By leveraging the distribution of input and output sequences, it effectively allocates resources and determines optimal execution configurations, including batch sizes and partial tensor parallelism. We also introduce two scheduling strategies based on Round-Robin Allocation and Workload-Aware Allocation policies, suitable for different NLP workloads. We evaluate ExeGPT on six LLM instances of T5, OPT, and GPT-3 and five NLP tasks, each with four distinct latency constraints. Compared to FasterTransformer, ExeGPT achieves up to 15.2x improvements in throughput and 6x improvements in latency. Overall, ExeGPT achieves an average throughput gain of 2.9x across twenty evaluation scenarios. Moreover, when adapting to changing sequence distributions, the cost of adjusting the schedule in ExeGPT is reasonably modest. ExeGPT proves to be an effective solution for optimizing and executing LLM inference for diverse NLP workload and serving conditions.

cs.DC↗

arXiv2023

IR-UWB Radar-Based Contactless Silent Speech Recognition of Vowels, Consonants, Words, and Phrases

Several sensing techniques have been proposed for silent speech recognition (SSR); however, many of these methods require invasive processes or sensor attachment to the skin using adhesive tape or glue, rendering them unsuitable for frequent use in daily life. By contrast, impulse radio ultra-wideband (IR-UWB) radar can operate without physical contact with users' articulators and related body parts, offering several advantages for SSR. These advantages include high range resolution, high penetrability, low power consumption, robustness to external light or sound interference, and the ability to be embedded in space-constrained handheld devices. This study demonstrated IR-UWB radar-based contactless SSR using four types of speech stimuli (vowels, consonants, words, and phrases). To achieve this, a novel speech feature extraction algorithm specifically designed for IR-UWB radar-based SSR is proposed. Each speech stimulus is recognized by applying a classification algorithm to the extracted speech features. Two different algorithms, multidimensional dynamic time warping (MD-DTW) and deep neural network-hidden Markov model (DNN-HMM), were compared for the classification task. Additionally, a favorable radar antenna position, either in front of the user's lips or below the user's chin, was determined to achieve higher recognition accuracy. Experimental results demonstrated the efficacy of the proposed speech feature extraction algorithm combined with DNN-HMM for classifying vowels, consonants, words, and phrases. Notably, this study represents the first demonstration of phoneme-level SSR using contactless radar.

eess.AS↗

arXiv2023

X-RiSAWOZ: High-Quality End-to-End Multilingual Dialogue Datasets and Few-shot Agents

Task-oriented dialogue research has mainly focused on a few popular languages like English and Chinese, due to the high dataset creation cost for a new language. To reduce the cost, we apply manual editing to automatically translated data. We create a new multilingual benchmark, X-RiSAWOZ, by translating the Chinese RiSAWOZ to 4 languages: English, French, Hindi, Korean; and a code-mixed English-Hindi language. X-RiSAWOZ has more than 18,000 human-verified dialogue utterances for each language, and unlike most multilingual prior work, is an end-to-end dataset for building fully-functioning agents. The many difficulties we encountered in creating X-RiSAWOZ led us to develop a toolset to accelerate the post-editing of a new language dataset after translation. This toolset improves machine translation with a hybrid entity alignment technique that combines neural with dictionary-based methods, along with many automated and semi-automated validation checks. We establish strong baselines for X-RiSAWOZ by training dialogue agents in the zero- and few-shot settings where limited gold data is available in the target language. Our results suggest that our translation and post-editing methodology and toolset can be used to create new high-quality multilingual dialogue agents cost-effectively. Our dataset, code, and toolkit are released open-source.

cs.CL↗

arXiv2023

Low-Cost GNSS Simulators with Wireless Clock Synchronization for Indoor Positioning

In regions where global navigation satellite systems (GNSS) signals are unavailable, such as underground areas and tunnels, GNSS simulators can be deployed for transmitting simulated GNSS signals. Then, a GNSS receiver in the simulator coverage outputs the position based on the received GNSS signals (e.g., Global Positioning System (GPS) L1 signals in this study) transmitted by the corresponding simulator. This approach provides periodic position updates to GNSS users while deploying a small number of simulators without modifying the hardware and software of user receivers. However, the simulator clock should be synchronized to the GNSS satellite clock to generate almost identical signals to the live-sky GNSS signals, which is necessary for seamless indoor and outdoor positioning handover. The conventional clock synchronization method based on the wired connection between each simulator and an outdoor GNSS antenna causes practical difficulty and increases the cost of deploying the simulators. This study proposes a wireless clock synchronization method based on a private time server and time delay calibration. Additionally, we derived the constraints for determining the optimal simulator coverage and separation between adjacent simulators. The positioning performance of the proposed GPS simulator-based indoor positioning system was demonstrated in the underground testbed for a driving vehicle with a GPS receiver and a pedestrian with a smartphone. The average position errors were 3.7 m for the vehicle and 9.6 m for the pedestrian during the field tests with successful indoor and outdoor positioning handovers. Since those errors are within the coverage of each deployed simulator, it is confirmed that the proposed system with wireless clock synchronization can effectively provide periodic position updates to users where live-sky GNSS signals are unavailable.

eess.SP↗

arXiv2023

Machine-Learning-Based Classification of GPS Signal Reception Conditions Using a Dual-Polarized Antenna in Urban Areas

In urban areas, dense buildings frequently block and reflect global positioning system (GPS) signals, resulting in the reception of a few visible satellites with many multipath signals. This is a significant problem that results in unreliable positioning in urban areas. If a signal reception condition from a certain satellite can be detected, the positioning performance can be improved by excluding or de-weighting the multipath contaminated satellite signal. Thus, we developed a machine-learning-based method of classifying GPS signal reception conditions using a dual-polarized antenna. We employed a decision tree algorithm for classification using three features, one of which can be obtained only from a dual-polarized antenna. A machine-learning model was trained using GPS signals collected from various locations. When the features extracted from the GPS raw signal are input, the generated machine-learning model outputs one of the three signal reception conditions: non-line-of-sight (NLOS) only, line-of-sight (LOS) only, or LOS+NLOS. Multiple testing datasets were used to analyze the classification accuracy, which was then compared with an existing method using dual single-polarized antennas. Consequently, when the testing dataset was collected at different locations from the training dataset, a classification accuracy of 64.47% was obtained, which was slightly higher than the accuracy of the existing method using dual single-polarized antennas. Therefore, the dual-polarized antenna solution is more beneficial than the dual single-polarized antenna solution because it has a more compact form factor and its performance is similar to that of the other solution.

cs.LG↗

arXiv2023

Performance Comparison of Numerical Optimization Algorithms for RSS-TOA-Based Target Localization

The maximum likelihood (ML) estimator can be applied to localize a target mobile device using the RSS and TOA. However, the ML estimator for the RSS-TOA-based target localization problem is nonconvex and nonlinear, having no analytical solution. Therefore, the ML estimator should be solved numerically, unless it is relaxed into a convex or linear form. This study investigates the target localization performance and computational complexity of numerical methods for solving an ML estimator. The three widely used numerical methods are: grid search, gradient descent, and particle swarm optimization. In the experimental evaluation, the grid search yielded the lowest target localization root-mean-squared error; however, the 95th percentile error of the grid search was larger than those of the other two algorithms. The average code computation time of the grid search was extremely large compared with those of the other two algorithms, and gradient descent exhibited the lowest computation time.

eess.SP↗

arXiv2023

Testing the Channels of Convolutional Neural Networks

Neural networks have complex structures, and thus it is hard to understand their inner workings and ensure correctness. To understand and debug convolutional neural networks (CNNs) we propose techniques for testing the channels of CNNs. We design FtGAN, an extension to GAN, that can generate test data with varying the intensity (i.e., sum of the neurons) of a channel of a target CNN. We also proposed a channel selection algorithm to find representative channels for testing. To efficiently inspect the target CNN's inference computations, we define unexpectedness score, which estimates how similar the inference computation of the test data is to that of the training data. We evaluated FtGAN with five public datasets and showed that our techniques successfully identify defective channels in five different CNN models.

cs.LG↗

arXiv2023

Performance Evaluation and Hybrid Application of the Greedy and Predictive UAV Trajectory Optimization Methods for Localizing a Target Mobile Device

This study investigates unmanned aerial vehicle (UAV) trajectory planning strategies for localizing a target mobile device in emergency situations. The global navigation satellite system (GNSS)-based accurate position information of a target mobile device in an emergency may not be always available to first responders. For example, 1) GNSS positioning accuracy may be degraded in harsh signal environments and 2) in countries where emergency positioning service is not mandatory, some mobile devices may not report their locations. Under the cases mentioned above, one way to find the target mobile device is to use UAVs. Dispatched UAVs may search the target directly on the emergency site by measuring the strength of the signal (e.g., LTE wireless communication signal) from the target mobile device. To accurately localize the target mobile device in the shortest time possible, UAVs should fly in the most efficient way possible. The two popular trajectory optimization strategies of UAVs are greedy and predictive approaches. However, the research on localization performances of the two approaches has been evaluated only under favorable settings (i.e., under good UAV geometries and small received signal strength (RSS) errors); more realistic scenarios still remain unexplored. In this study, we compare the localization performance of the greedy and predictive approaches under realistic RSS errors (i.e., up to 6 dB according to the ITU-R channel model).

eess.SP↗

arXiv2022

Sample-Efficient Training of Robotic Guide Using Human Path Prediction Network

Training a robot that engages with people is challenging; it is expensive to directly involve people in the training process, which requires numerous data samples. This paper presents an alternative approach for resolving this problem. We propose a human path prediction network (HPPN) that generates a user's future trajectory based on sequential robot actions and human responses using a recurrent-neural-network structure. Subsequently, an evolution-strategy-based robot training method using only the virtual human movements generated using the HPPN is presented. It is demonstrated that our proposed method permits sample-efficient training of a robotic guide for visually impaired people. By collecting only 1.5 K episodes from real users, we were able to train the HPPN and generate more than 100 K virtual episodes required for training the robot. The trained robot precisely guided blindfolded participants along a target path. Furthermore, using virtual episodes, we investigated a new reward design that prioritizes human comfort during the robot's guidance without incurring additional costs. This sample-efficient training method is expected to be widely applicable to future robots that interact physically with humans.

cs.RO↗

arXiv2022

Urban Road Safety Prediction: A Satellite Navigation Perspective

Predicting the safety of urban roads for navigation via global navigation satellite systems (GNSS) signals is considered. To ensure safe driving of automated vehicles, the vehicle must plan its trajectory to avoid navigating on unsafe roads (e.g., icy conditions, construction zones, narrow streets, etc.). Such information can be derived from the roads' physical properties, vehicle's capabilities, and weather conditions. From a GNSS-based navigation perspective, the reliability of GNSS signals in different locales, which is heavily dependent on the road layout within the surrounding environment, is crucial to ensure safe automated driving. An urban road environment surrounded by tall objects can significantly degrade the accuracy and availability of GNSS signals. This article proposes an approach to predict the reliability of GNSS-based navigation to ensure safe urban navigation. Satellite navigation reliability at a given location and time on a road is determined based on the probabilistic position error bound of the vehicle-mounted GNSS receiver. A metric for GNSS reliability for ground vehicles is suggested, and a method to predict the conservative probabilistic error bound of the GNSS navigation solution is proposed. A satellite navigation reliability map is generated for various navigation applications. As a case study, the reliability map is used in the proposed optimization problem formulation for automated ground vehicle safety-constrained path planning.

eess.SP↗

arXiv2022

Evaluation of RF Fingerprinting-Aided RSS-Based Target Localization for Emergency Response

Target localization is essential for emergency dispatching situations. Maximum likelihood estimation (MLE) methods are widely used to estimate the target position based on the received signal strength measurements. However, the performance of MLE solvers is significantly affected by the initialization (i.e., initial guess of the solution or solution search space). To address this, a previous study proposed the semidefinite programming (SDP)-based MLE initialization. However, the performance of the SDP-based initialization technique is largely affected by the shadowing variance and geometric diversity between the target and receivers. In this study, a radio frequency (RF) fingerprinting-based MLE initialization is proposed. Further, a maximum likelihood problem for target localization combining RF fingerprinting is formulated. In the three test environments of open space, urban, and indoor, the proposed RF fingerprinting-aided target localization method showed a performance improvement of up to 63.31% and an average of 39.13%, compared to the MLE algorithm initialized with SDP. Furthermore, unlike the SDP-MLE method, the proposed method was not significantly affected by the poor geometry between the target and receivers in our experiments.

eess.SP↗

arXiv2022

Seamless Accurate Positioning in Deep Urban Area based on Mode Switching Between DGNSS and Multipath Mitigation Positioning

Multipath and non-line-of-sight (NLOS) signals are the major causes of poor accuracy of a global navigation satellite system (GNSS) in urban areas. Despite the wide usage of the GNSS in populated urban areas, it is difficult to suggest a generalized method because multipath errors are user-specific errors that cannot be eliminated by the DGNSS or a real-time kinematic technique. This paper introduces a real-time multipath estimation and mitigation technique, which considers compensation for the time offset between constellations. It also presents a mode-switching algorithm between the DGNSS and multipath mitigating mode and shows that this technique can be effectively utilized for automobiles in a deep urban environment without any help from sensors other than GNSS. The availability is improved from 64% to 100% and the error RMS is reduced from 11.1 m to 1.2 m on Teheran-ro, Seoul, Korea. Because this method does not require prior information or additional sensor implementation for high-positioning performance in deep urban areas, it is expected to gain wide usage in not only the automotive industry but also future intelligent transportation systems.

eess.SY↗

arXiv2022

Optimal Parameter Inflation to Enhance the Availability of Single-Frequency GBAS for Intelligent Air Transportation

Ground-based Augmentation System (GBAS) augments Global Navigation Satellite Systems (GNSS) to support the precision approach and landing of aircraft. To guarantee integrity, existing single-frequency GBAS utilizes position-domain geometry screening to eliminate potentially unsafe satellite geometries by inflating one or more broadcast GBAS parameters. However, GBAS availability can be drastically impacted in low-latitude regions where severe ionospheric conditions have been observed. Thus, we developed a novel geometry-screening algorithm in this study to improve GBAS availability in low-latitude regions. Simulations demonstrate that the proposed method can provide 5-8 percentage point availability enhancement of GBAS at Galeão airport near Rio de Janeiro, Brazil, compared to existing methods.

eess.SY↗

arXiv2021

SFOL DME Pulse Shaping Through Digital Predistortion for High-Accuracy DME

The Stretched-FrOnt-Leg (SFOL) pulse is a high-accuracy distance measuring equipment (DME) pulse developed to support alternative positioning and navigation for aircraft during global navigation satellite system outages. To facilitate the use of the SFOL pulse, it is best to use legacy DMEs that are already deployed to transmit the SFOL pulse, rather than the current Gaussian pulse, through software changes only. When attempting to transmit the SFOL pulse in legacy DMEs, the greatest challenge is the pulse shape distortion caused by the pulse-shaping circuits and power amplifiers in the transmission unit such that the original SFOL pulse shape is no longer preserved. This letter proposes an inverse-learning-based DME digital predistortion method and presents successfully transmitted SFOL pulses from a testbed based on a commercial legacy DME that was designed to transmit Gaussian pulses.

eess.SY↗

arXiv2021

Fast User Adaptation for Human Motion Prediction in Physical Human-Robot Interaction

Accurate prediction of human movements is required to enhance the efficiency of physical human-robot interaction. Behavioral differences across various users are crucial factors that limit the prediction of human motion. Although recent neural network-based modeling methods have improved their prediction accuracy, most did not consider an effective adaptations to different users, thereby employing the same model parameters for all users. To deal with this insufficiently addressed challenge, we introduce a meta-learning framework to facilitate the rapid adaptation of the model to unseen users. In this study, we propose a model structure and a meta-learning algorithm specialized to enable fast user adaptation in predicting human movements in cooperative situations with robots. The proposed prediction model comprises shared and adaptive parameters, each addressing the user's general and individual movements. Using only a small amount of data from an individual user, the adaptive parameters are adjusted to enable user-specific prediction through a two-step process: initialization via a separate network and adaptation via a few gradient steps. Regarding the motion dataset that has 20 users collaborating with a robotic device, the proposed method outperforms existing meta-learning and non-meta-learning baselines in predicting the movements of unseen users.

cs.RO↗

arXiv2021

Scheduling Optimization Techniques for Neural Network Training

Neural network training requires a large amount of computation and thus GPUs are often used for the acceleration. While they improve the performance, GPUs are underutilized during the training.This paper proposes out-of-order (ooo) backprop, an effective scheduling technique for neural network training. By exploiting the dependencies of gradient computations, ooo backprop enables to reorder their executions to make the most of the GPU resources. We show that the GPU utilization in single-GPU, data-parallel, and pipeline-parallel training can be commonly improve by applying ooo back-prop and prioritizing critical operations. We propose three scheduling algorithms based on ooo backprop. For single-GPU training, we schedule with multi-stream out-of-order computation to mask the kernel launch overhead. In data-parallel training, we reorder the gradient computations to maximize the overlapping of computation and parameter communication; in pipeline-parallel training, we prioritize critical gradient computations to reduce the pipeline stalls.We evaluate our optimizations with twelve neural networks including a light-weight computer vision model (MobileNet) and largeNLP models (BERT and GPT-3) with up to forty eight V100 GPUs.Our scheduling algorithms effectively improve the performance of single-GPU training as well as data- and pipeline-parallel training.Compared to the respective state of the art training systems, the throughput is substantially improved for single-GPU, data-parallel, and pipeline-parallel training.

cs.LG↗

arXiv2021

First Demonstration of the Korean eLoran Accuracy in a Narrow Waterway Using Improved ASF Maps

The vulnerabilities of global navigation satellite systems (GNSSs) to radio frequency jamming and spoofing have attracted significant research attention. In particular, the large-scale jamming incidents that occurred in South Korea substantiate the practical importance of implementing a complementary navigation system. This letter briefly summarizes the efforts of South Korea to deploy an enhanced long-range navigation (eLoran) system, which is a terrestrial low-frequency radio navigation system that can complement GNSSs. After four years of research and development, the Korean eLoran testbed system has been recently deployed and is operational since June 1, 2021. Although its initial performance at sea is satisfactory, navigation through a narrow waterway is still challenging because a complete survey of the additional secondary factor (ASF), which is the largest source of error for eLoran, is practically difficult in a narrow waterway. This letter proposes an alternative way to survey the ASF in a narrow waterway and improve the ASF map generation methods. Moreover, the performance of the proposed approach was validated experimentally.

eess.SP↗