arXiv subjects

Feng Guo

Publications and source records attributed to Feng Guo.

At least 73 records · Page 4Linked to original sources

arXiv2022

PL-VINS: Real-Time Monocular Visual-Inertial SLAM with Point and Line Features

Leveraging line features to improve localization accuracy of point-based visual-inertial SLAM (VINS) is gaining interest as they provide additional constraints on scene structure. However, real-time performance when incorporating line features in VINS has not been addressed. This paper presents PL-VINS, a real-time optimization-based monocular VINS method with point and line features, developed based on the state-of-the-art point-based VINS-Mono \cite{vins}. We observe that current works use the LSD \cite{lsd} algorithm to extract line features; however, LSD is designed for scene shape representation instead of the pose estimation problem, which becomes the bottleneck for the real-time performance due to its high computational cost. In this paper, a modified LSD algorithm is presented by studying a hidden parameter tuning and length rejection strategy. The modified LSD can run at least three times as fast as LSD. Further, by representing space lines with the Plücker coordinates, the residual error in line estimation is modeled in terms of the point-to-line distance, which is then minimized by iteratively updating the minimum four-parameter orthonormal representation of the Plücker coordinates. Experiments in a public benchmark dataset show that the localization error of our method is 12-16\% less than that of VINS-Mono at the same pose update frequency. %For the benefit of the community, The source code of our method is available at: https://github.com/cnqiangfu/PL-VINS.

cs.RO↗

arXiv2022

Limits of real bivariate rational functions

Given two nonzero polynomials $f, g \in\mathbb R[x,y]$ and a point $(a, b) \in \mathbb{R}^2,$ we give some necessary and sufficient conditions for the existence of the limit $\displaystyle \lim_{(x, y) \to (a, b)} \frac{f(x, y)}{g(x, y)}.$ We also show that, if the denominator $g$ has an isolated zero at the given point $(a, b),$ then the set of possible limits of $\displaystyle \lim_{(x, y) \to (a, b)} \frac{f(x, y)}{g(x, y)}$ is a closed interval in $\overline{\mathbb{R}}$ and can be explicitly determined. As an application, we propose an effective algorithm to verify the existence of the limit and compute the limit (if it exists). Our approach is geometric and is based on Puiseux expansions.

math.CA↗

arXiv2022

Smartphone-based Hard-braking Event Detection at Scale for Road Safety Services

Road crashes are the sixth leading cause of lost disability-adjusted life-years (DALYs) worldwide. One major challenge in traffic safety research is the sparsity of crashes, which makes it difficult to achieve a fine-grain understanding of crash causations and predict future crash risk in a timely manner. Hard-braking events have been widely used as a safety surrogate due to their relatively high prevalence and ease of detection with embedded vehicle sensors. As an alternative to using sensors fixed in vehicles, this paper presents a scalable approach for detecting hard-braking events using the kinematics data collected from smartphone sensors. We train a Transformer-based machine learning model for hard-braking event detection using concurrent sensor readings from smartphones and vehicle sensors from drivers who connect their phone to the vehicle while navigating in Google Maps. The detection model shows superior performance with a $0.83$ Area under the Precision-Recall Curve (PR-AUC), which is $3.8\times$better than a GPS speed-based heuristic model, and $166.6\times$better than an accelerometer-based heuristic model. The detected hard-braking events are strongly correlated with crashes from publicly available datasets, supporting their use as a safety surrogate. In addition, we conduct model fairness and selection bias evaluation to ensure that the safety benefits are equally shared. The developed methodology can benefit many safety applications such as identifying safety hot spots at road network level, evaluating the safety of new user interfaces, as well as using routing to improve traffic safety.

cs.LG↗

arXiv2021

On Solving a Class of Fractional Semi-infinite Polynomial Programming Problems

In this paper, we study a class of fractional semi-infinite polynomial programming (FSIPP) problems, in which the objective is a fraction of a convex polynomial and a concave polynomial, and the constraints consist of infinitely many convex polynomial inequalities. To solve such a problem, we first reformulate it to a pair of primal and dual conic optimization problems, which reduce to semidefinite programming (SDP) problems if we can bring sum-of-squares structures into the conic constraints. To this end, we provide a characteristic cone constraint qualification for convex semi-infinite programming problems to guarantee strong duality and also the attainment of the solution in the dual problem, which is of its own interest. In this framework, we first present a hierarchy of SDP relaxations with asymptotic convergence for the FSIPP problem whose index set is defined by finitely many polynomial inequalities. Next, we study four cases of the FSIPP problems which can be reduced to either a single SDP problem or a finite sequence of SDP problems, where at least one minimizer can be extracted. Then, we apply this approach to the four corresponding multi-objective cases to find efficient solutions.

math.OC↗

arXiv2021

Greenplum: A Hybrid Database for Transactional and Analytical Workloads

Demand for enterprise data warehouse solutions to support real-time Online Transaction Processing (OLTP) queries as well as long-running Online Analytical Processing (OLAP) workloads is growing. Greenplum database is traditionally known as an OLAP data warehouse system with limited ability to process OLTP workloads. In this paper, we augment Greenplum into a hybrid system to serve both OLTP and OLAP workloads. The challenge we address here is to achieve this goal while maintaining the ACID properties with minimal performance overhead. In this effort, we identify the engineering and performance bottlenecks such as the under-performing restrictive locking and the two-phase commit protocol. Next we solve the resource contention issues between transactional and analytical queries. We propose a global deadlock detector to increase the concurrency of query processing. When transactions that update data are guaranteed to reside on exactly one segment we introduce one-phase commit to speed up query processing. Our resource group model introduces the capability to separate OLAP and OLTP workloads into more suitable query processing mode. Our experimental evaluation on the TPC-B and CH-benCHmark benchmarks demonstrates the effectiveness of our approach in boosting the OLTP performance without sacrificing the OLAP performance.

cs.DB↗

arXiv2021

Global Łojasiewicz inequalities on comparing the rate of growth of polynomial functions

We present a global version of the Łojasiewicz inequality on comparing the rate of growth of two polynomial functions in the case the mapping defined by these functions is (Newton) non-degenerate at infinity. In addition, we show that the condition of non-degeneracy at infinity is generic in the sense that it holds in an open and dense semi-algebraic set of the entire space of input data.

math.AG↗

arXiv2020

The Devil is in the Details: Delving into Unbiased Data Processing for Human Pose Estimation

Being a fundamental component in training and inference, data processing has not been systematically considered in human pose estimation community, to the best of our knowledge. In this paper, we focus on this problem and find that the devil of human pose estimation evolution is in the biased data processing. Specifically, by investigating the standard data processing in state-of-the-art approaches mainly including coordinate system transformation and keypoint format transformation (i.e., encoding and decoding), we find that the results obtained by common flipping strategy are unaligned with the original ones in inference. Moreover, there is a statistical error in some keypoint format transformation methods. Two problems couple together, significantly degrade the pose estimation performance and thus lay a trap for the research community. This trap has given bone to many suboptimal remedies, which are always unreported, confusing but influential. By causing failure in reproduction and unfair in comparison, the unreported remedies seriously impedes the technological development. To tackle this dilemma from the source, we propose Unbiased Data Processing (UDP) consist of two technique aspect for the two aforementioned problems respectively (i.e., unbiased coordinate system transformation and unbiased keypoint format transformation). As a model-agnostic approach and a superior solution, UDP successfully pushes the performance boundary of human pose estimation and offers a higher and more reliable baseline for research community. Code is public available in https://github.com/HuangJunJie2017/UDP-Pose

cs.CV↗

arXiv2020

On continuous selections of polynomial functions

A continuous selection of polynomial functions is a continuous function whose domain can be partitioned into finitely many pieces on which the function coincides with a polynomial. Given a set of finitely many polynomials, we show that there are only finitely many continuous selections of it and each one is semi-algebraic. Then, we establish some generic properties regarding the critical points, defined by the Clarke subdifferential, of these continuous selections. In particular, given a set of finitely many polynomials with generic coefficients, we show that the critical points of all continuous selections of it are finite and the critical values are all different, and we also derive the coercivity of those continuous selections which are bounded from below. We point out that some existing results about Łojasiewicz's inequality and error bounds for the maximum function of some finitely many polynomials are also valid for all the continuous selections of them.

math.OC↗

arXiv2020

Salience and Market-aware Skill Extraction for Job Targeting

At LinkedIn, we want to create economic opportunity for everyone in the global workforce. To make this happen, LinkedIn offers a reactive Job Search system, and a proactive Jobs You May Be Interested In (JYMBII) system to match the best candidates with their dream jobs. One of the most challenging tasks for developing these systems is to properly extract important skill entities from job postings and then target members with matched attributes. In this work, we show that the commonly used text-based \emph{salience and market-agnostic} skill extraction approach is sub-optimal because it only considers skill mention and ignores the salient level of a skill and its market dynamics, i.e., the market supply and demand influence on the importance of skills. To address the above drawbacks, we present \model, our deployed \emph{salience and market-aware} skill extraction system. The proposed \model ~shows promising results in improving the online performance of job recommendation (JYMBII) ($+1.92\%$ job apply) and skill suggestions for job posters ($-37\%$ suggestion rejection rate). Lastly, we present case studies to show interesting insights that contrast traditional skill recognition method and the proposed \model~from occupation, industry, country, and individual skill levels. Based on the above promising results, we deployed the \model ~online to extract job targeting skills for all $20$M job postings served at LinkedIn.

cs.IR↗

arXiv2020

On types of KKT points in polynomial optimization

Let $f$ be a real polynomial function with $n$ variables and $S$ be a basic closed semialgebraic set in $\Bbb{R}^n$. In this paper, we are interested in the problem of identifying the type (local minimizer, maximizer or not extremum point) of a given isolated KKT point $x^*$ of $f$ over $S.$ To this end, we investigate some properties of the tangency variety of $f$ on $S$ at $x^*,$ by which we introduce the definition of faithful radius of $f$ over $S$ at $x^*.$ Then, we show that the type of $x^*$ can be determined by the global extrema of $f$ over the intersection of $S$ and the Euclidean ball centered at $x^*$ with a faithful radius. Finally, we propose an algorithm involving algebraic computations to compute a faithful radius of $x^*$ and determine its type.

math.OC↗

arXiv2020

AutoQ: Automated Kernel-Wise Neural Network Quantization

Network quantization is one of the most hardware friendly techniques to enable the deployment of convolutional neural networks (CNNs) on low-power mobile devices. Recent network quantization techniques quantize each weight kernel in a convolutional layer independently for higher inference accuracy, since the weight kernels in a layer exhibit different variances and hence have different amounts of redundancy. The quantization bitwidth or bit number (QBN) directly decides the inference accuracy, latency, energy and hardware overhead. To effectively reduce the redundancy and accelerate CNN inferences, various weight kernels should be quantized with different QBNs. However, prior works use only one QBN to quantize each convolutional layer or the entire CNN, because the design space of searching a QBN for each weight kernel is too large. The hand-crafted heuristic of the kernel-wise QBN search is so sophisticated that domain experts can obtain only sub-optimal results. It is difficult for even deep reinforcement learning (DRL) Deep Deterministic Policy Gradient (DDPG)-based agents to find a kernel-wise QBN configuration that can achieve reasonable inference accuracy. In this paper, we propose a hierarchical-DRL-based kernel-wise network quantization technique, AutoQ, to automatically search a QBN for each weight kernel, and choose another QBN for each activation layer. Compared to the models quantized by the state-of-the-art DRL-based schemes, on average, the same models quantized by AutoQ reduce the inference latency by 54.06\%, and decrease the inference energy consumption by 50.69\%, while achieving the same inference accuracy.

cs.LG↗

arXiv2020

Multitasking additional-to-driving: Prevalence, structure, and associated risk in SHRP2 naturalistic driving data

This paper 1) analyzes the extent to which drivers engage in multitasking additional-to-driving (MAD) under various conditions, 2) specifies odds ratios (ORs) of crashing associated with MAD compared to no task engagement, and 3) explores the structure of MAD, based on data from the Second Strategic Highway Research Program Naturalistic Driving Study (SHRP2 NDS). Sensitivity analysis in which secondary tasks were re-defined by grouping similar tasks was performed to investigate the extent to which ORs are affected by the specific task definitions in SHRP2. A novel visual representation of multitasking was developed to show which secondary tasks co-occur frequently and which ones do not. MAD occurs in 11% of control driving segments, 22% of crashes and near-crashes (CNC), 26% of Level 1-3 crashes and 39% of rear-end striking crashes, and 9%, 16%, 17% and 28% respectively for the same event types if MAD is defined in terms of general task groups. The most common co-occurrences of secondary tasks vary substantially among event types; for example, 'Passenger in adjacent seat - interaction' and 'Other non-specific internal eye glance' tend to co-occur in CNC but tend not to co-occur in control driving segments. The odds ratios of MAD compared to driving without any secondary task and the corresponding 95% confidence intervals are 2.38 (2.17-2.61) for CNC, 3.72 (3.11-4.45) for Level 1-3 crashes and 8.48 (5.11-14.07) for rear-end striking crashes. The corresponding ORs using general task groups to define MAD are slightly lower at 2.00 (1.80-2.21) for CNC, 3.03 (2.48-3.69) for Level 1-3 crashes and 6.94 (4.04-11.94) for rear-end striking crashes. The results confirm that independently of whether secondary tasks are defined according to SHRP2 or general task groups, the reduction of driving performance from MAD observed in simulator studies is manifested in real-world crashes as well.

stat.AP↗

arXiv2020

MindReading: An Ultra-Low-Power Photonic Accelerator for EEG-based Human Intention Recognition

A scalp-recording electroencephalography (EEG)-based brain-computer interface (BCI) system can greatly improve the quality of life for people who suffer from motor disabilities. Deep neural networks consisting of multiple convolutional, LSTM and fully-connected layers are created to decode EEG signals to maximize the human intention recognition accuracy. However, prior FPGA, ASIC, ReRAM and photonic accelerators cannot maintain sufficient battery lifetime when processing real-time intention recognition. In this paper, we propose an ultra-low-power photonic accelerator, MindReading, for human intention recognition by only low bit-width addition and shift operations. Compared to prior neural network accelerators, to maintain the real-time processing throughput, MindReading reduces the power consumption by 62.7\% and improves the throughput per Watt by 168\%.

eess.SP↗

arXiv2019

Semidefinite programming relaxations for linear semi-infinite polynomial programming

This paper studies a class of so-called linear semi-infinite polynomial programming (LSIPP) problems. It is a subclass of linear semi-infinite programming problems whose constraint functions are polynomials in parameters and index sets are basic semialgebraic sets. We present a hierarchy of semidefinite programming (SDP) relaxations for LSIPP problems. Convergence rate analysis of the SDP relaxations is established based on some existing results. We show how to verify the compactness of feasible sets of LSIPP problems. In the end, we extend the SDP relaxation method to more general semi-infinite programming problems.

math.OC↗

arXiv2019

Semi-Supervised Adversarial Monocular Depth Estimation

In this paper, we address the problem of monocular depth estimation when only a limited number of training image-depth pairs are available. To achieve a high regression accuracy, the state-of-the-art estimation methods rely on CNNs trained with a large number of image-depth pairs, which are prohibitively costly or even infeasible to acquire. Aiming to break the curse of such expensive data collections, we propose a semi-supervised adversarial learning framework that only utilizes a small number of image-depth pairs in conjunction with a large number of easily-available monocular images to achieve high performance. In particular, we use one generator to regress the depth and two discriminators to evaluate the predicted depth , i.e., one inspects the image-depth pair while the other inspects the depth channel alone. These two discriminators provide their feedbacks to the generator as the loss to generate more realistic and accurate depth predictions. Experiments show that the proposed approach can (1) improve most state-of-the-art models on the NYUD v2 dataset by effectively leveraging additional unlabeled data sources; (2) reach state-of-the-art accuracy when the training set is small, e.g., on the Make3D dataset; (3) adapt well to an unseen new dataset (Make3D in our case) after training on an annotated dataset (KITTI in our case).

cs.CV↗

arXiv2019

On semi-infinite systems of convex polynomial inequalities and polynomial

We consider the semi-infinite system of polynomial inequalities of the form \[ \mathbf{K}:=\{x\in\mathbb{R}^m\mid p(x,y)\ge 0,\ \ \forall y\in S\subseteq\mathbb{R}^n\}, \] where $p(x,y)$ is a real polynomial in the variables $x$ and the parameters $y$, the index set $S$ is a basic semialgebraic set in $\mathbb{R}^n$, $-p(x,y)$ is convex in $x$ for every $y\in S$. We propose a procedure to construct approximate semidefinite representations of $\mathbf{K}$. There are two indices to index these approximate semidefinite representations. As two indices increase, these semidefinite representation sets expand and contract, respectively, and can approximate $\mathbf{K}$ as closely as possible under some assumptions. In some special cases, we can fix one of the two indices or both. Then, we consider the optimization problem of minimizing a convex polynomial over $\mathbf{K}$. We present an SDP relaxation method for this optimization problem by similar strategies used in constructing approximate semidefinite representations of $\mathbf{K}$. Under certain assumptions, some approximate minimizers of the optimization problem can also be obtained from the SDP relaxations. In some special cases, we show that the SDP relaxation for the optimization problem is exact and all minimizers can be extracted.

math.OC↗

arXiv2017

On types of degenerate critical points of real polynomial functions

In this paper, we consider the problem of identifying the type (local minimizer, maximizer or saddle point) of a given isolated real critical point $c$, which is degenerate, of a multivariate polynomial function $f$. To this end, we introduce the definition of faithful radius of $c$ by means of the curve of tangency of $f$. We show that the type of $c$ can be determined by the global extrema of $f$ over the Euclidean ball centered at $c$ with a faithful radius.We propose algorithms to compute a faithful radius of $c$ and determine its type.

math.OC↗

arXiv2017

Model of ultrafast demagnetization driven by spin-orbit coupling in a photoexcited antiferromagnetic insulator Cr2O3

We theoretically study the dynamic time evolution following laser pulse pumping in an antiferromagnetic insulator Cr$_{2}$O$_{3}$. From the photoexcited high-spin quartet states to the long-lived low-spin doublet states, the ultrafast demagnetization processes are investigated by solving the dissipative Schrödinger equation. We find that the demagnetization times are of the order of hundreds of femtosecond, in good agreement with recent experiments. The switching times could be strongly reduced by properly tuning the energy gaps between the multiplet energy levels of Cr$^{3+}$. Furthermore, the relaxation times also depend on the hybridization of atomic orbitals in the first photoexcited state. Our results suggest that the selective manipulation of electronic structure by engineering stress-strain or chemical substitution allows effective control of the magnetic state switching in photoexcited insulating transition-metal oxides.

cond-mat.str-el↗