arXiv subjects

Ru Wang

Publications and source records attributed to Ru Wang.

33 records · Page 2Linked to original sources

arXiv2023

Peer attention enhances student learning

Human visual attention is susceptible to social influences. In education, peer effects impact student learning, but their precise role in modulating attention remains unclear. Our experiment (N=311) demonstrates that displaying peer visual attention regions when students watch online course videos enhances their focus and engagement. However, students retain adaptability in following peer attention cues. Overall, guided peer attention improves learning experiences and outcomes. These findings elucidate how peer visual attention shapes students' gaze patterns, deepening understanding of peer influence on learning. They also offer insights into designing adaptive online learning interventions leveraging peer attention modelling to optimize student attentiveness and success.

cs.HC↗

arXiv2023

On Committor Functions in Milestoning

As an optimal one-dimensional reaction coordinate, the committor function not only describes the probability of a trajectory initiated at a phase space point first reaching the product state before reaching the reactant state, but also preserves the kinetics when utilized to run a reduced dynamics model. However, calculating the committor function in high-dimensional systems poses significant challenges. In this paper, within the framework of Milestoning, exact expressions for committor functions at two levels of coarse graining are given, including committor functions of phase space point to point (CFPP) and milestone to milestone (CFMM). When combined with transition kernels obtained from trajectory analysis, these expressions can be utilized to accurately and efficiently compute the committor functions. Furthermore, based on the calculated committor functions, an adaptive algorithm is developed to gradually refine the transition state region. Finally, two model examples are employed to assess the accuracy of these different formulations of committor functions.

physics.chem-ph↗

arXiv2023

Characterizing Barriers and Technology Needs in the Kitchen for Blind and Low Vision People

Cooking is a vital yet challenging activity for people with visual impairments (PVI). It involves tasks that can be dangerous or difficult without vision, such as handling a knife or adding a suitable amount of salt. A better understanding of these challenges can inform the design of technologies that mitigate safety hazards and improve the quality of the lives of PVI. Furthermore, there is a need to understand the effects of different visual abilities, including low vision and blindness, and the role of rehabilitation training where PVI learn cooking skills and assistive technologies. In this paper, we aim to comprehensively characterize PVI's challenges, strategies, and needs in the kitchen from the perspectives of both PVI and rehabilitation professionals. Through a contextual inquiry study, we observed 10 PVI, including six low vision and four blind participants, when they cooked dishes of their choices in their own kitchens. We then interviewed six rehabilitation professionals to explore their training strategies and technology recommendations. Our findings revealed the differences between low vision and blind people during cooking as well as the gaps between training and reality. We suggest improvements for rehabilitation training and distill design considerations for future assistive technology in the kitchen.

cs.HC↗

arXiv2023

Approximating First Hitting Point Distribution in Milestoning for Rare Event Kinetics

Milestoning is an efficient method for rare event kinetics calculation using short trajectory parallelization. Mean first passage time (MFPT) is the key kinetic output of Milestoning, whose accuracy crucially depends the initial distribution of the short trajectory ensemble. The true initial distribution, i.e., first hitting point distribution (FHPD), has no analytic expression in the general case. Here, we introduce two algorithms, local passage time weighted Milestoning (LPT-M) and Bayesian inference Milestoning (BI-M), to accurately and efficiently approximate FHPD for systems at equilibrium condition. Starting from sampling Boltzmann distribution on milestones, we calculate the proper weighting factor for the short trajectory ensemble. The methods are tested on two model examples for illustration purpose. Both methods improve significantly over the widely used classical Milestoning method in terms of the accuracy of MFPT. In particular, BI-M covers the directional Milestoning method as a special case in the deterministic Hamiltonian dynamics. LPT-M is especially advantageous in terms of computational costs and robustness with respect to the increasing number of intermediate milestones. Furthermore, a locally iterative correction algorithm for non-equilibrium stationary FHPD is developed for exact MFPT calculation, which can be combined with LPT-M/BI-M and is much cheaper than the exact Milestoning method.

physics.chem-ph↗

arXiv2023

Convergence of least energy sign-changing solutions for logarithmic Schrödinger equations on locally finite graphs

In this paper, we study the following logarithmic Schrödinger equation \[ -Δu+λa(x)u=u\log u^2\ \ \ \ \mbox{ in }V \] on a connected locally finite graph $G=(V,E)$, where $Δ$ denotes the graph Laplacian, $λ> 0$ is a constant, and $a(x) \geq 0$ represents the potential. Using variational techniques in combination with the Nehari manifold method based on directional derivative, we can prove that, there exists a constant $λ_0>0$ such that for all $λ\geqλ_0$, the above problem admits a least energy sign-changing solution $u_λ$. Moreover, as $λ\to+\infty$, we prove that the solution $u_λ$ converges to a least energy sign-changing solution of the following Dirichlet problem \[\begin{cases} -Δu=u\log u^2~~~&\mbox{ in }Ω,\\ u(x)=0~~~&\mbox{ on }\partialΩ, \end{cases}\] where $Ω=\{x\in V: a(x)=0\}$ is the potential well.

math.AP↗

arXiv2023

Understanding How Low Vision People Read Using Eye Tracking

While being able to read with screen magnifiers, low vision people have slow and unpleasant reading experiences. Eye tracking has the potential to improve their experience by recognizing fine-grained gaze behaviors and providing more targeted enhancements. To inspire gaze-based low vision technology, we investigate the suitable method to collect low vision users' gaze data via commercial eye trackers and thoroughly explore their challenges in reading based on their gaze behaviors. With an improved calibration interface, we collected the gaze data of 20 low vision participants and 20 sighted controls who performed reading tasks on a computer screen; low vision participants were also asked to read with different screen magnifiers. We found that, with an accessible calibration interface and data collection method, commercial eye trackers can collect gaze data of comparable quality from low vision and sighted people. Our study identified low vision people's unique gaze patterns during reading, building upon which, we propose design implications for gaze-based low vision technology.

cs.HC↗

arXiv2022

Ground states for logarithmic Schrödinger equations on locally finite graphs

In this paper, we study the following logarithmic Schrödinger equation \[ -Δu+a(x)u=u\log u^2\ \ \ \ \mbox{in }V, \] where $Δ$ is the graph Laplacian, $G=(V,E)$ is a connected locally finite graph, the potential $a: V\to \mathbb{R}$ is bounded from below and may change sign. We first establish two Sobolev compact embedding theorems in the case when different assumptions are imposed on $a(x)$. It leads to two kinds of associated energy functionals, one of which is not well-defined under the logarithmic nonlinearity, while the other is $C^1$. The existence of ground state solutions are then obtained by using the Nehari manifold method and the mountain pass theorem respectively.

math.AP↗

arXiv2022

NTIRE 2021 Challenge on Quality Enhancement of Compressed Video: Methods and Results

This paper reviews the first NTIRE challenge on quality enhancement of compressed video, with a focus on the proposed methods and results. In this challenge, the new Large-scale Diverse Video (LDV) dataset is employed. The challenge has three tracks. Tracks 1 and 2 aim at enhancing the videos compressed by HEVC at a fixed QP, while Track 3 is designed for enhancing the videos compressed by x265 at a fixed bit-rate. Besides, the quality enhancement of Tracks 1 and 3 targets at improving the fidelity (PSNR), and Track 2 targets at enhancing the perceptual quality. The three tracks totally attract 482 registrations. In the test phase, 12 teams, 8 teams and 11 teams submitted the final results of Tracks 1, 2 and 3, respectively. The proposed methods and solutions gauge the state-of-the-art of video quality enhancement. The homepage of the challenge: https://github.com/RenYang-home/NTIRE21_VEnh

eess.IV↗

arXiv2020

Learning Not to Learn in the Presence of Noisy Labels

Learning in the presence of label noise is a challenging yet important task: it is crucial to design models that are robust in the presence of mislabeled datasets. In this paper, we discover that a new class of loss functions called the gambler's loss provides strong robustness to label noise across various levels of corruption. We show that training with this loss function encourages the model to "abstain" from learning on the data points with noisy labels, resulting in a simple and effective method to improve robustness and generalization. In addition, we propose two practical extensions of the method: 1) an analytical early stopping criterion to approximately stop training before the memorization of noisy labels, as well as 2) a heuristic for setting hyperparameters which do not require knowledge of the noise corruption rate. We demonstrate the effectiveness of our method by achieving strong results across three image and text classification tasks as compared to existing baselines.

cs.LG↗

arXiv2018

Approximate Random Dropout

The training phases of Deep neural network~(DNN) consumes enormous processing time and energy. Compression techniques utilizing the sparsity of DNNs can effectively accelerate the inference phase of DNNs. However, it can be hardly used in the training phase because the training phase involves dense matrix-multiplication using General Purpose Computation on Graphics Processors (GPGPU), which endorse regular and structural data layout. In this paper, we propose the Approximate Random Dropout that replaces the conventional random dropout of neurons and synapses with a regular and predefined patterns to eliminate the unnecessary computation and data access. To compensate the potential performance loss we develop a SGD-based Search Algorithm to produce the distribution of dropout patterns. We prove our approach is statistically equivalent to the previous dropout method. Experiments results on MLP and LSTM using well-known benchmarks show that the proposed Approximate Random Dropout can reduce the training time by $20\%$-$77\%$ ($19\%$-$60\%$) when dropout rate is $0.3$-$0.7$ on MLP (LSTM) with marginal accuracy drop.

cs.LG↗

arXiv2015

Is it possible to formulate least action principle for dissipative systems?

A longstanding open question in classical mechanics is to formulate the least action principle for dissipative systems. In this work, we give a general formulation of this principle by considering a whole conservative system including the damped moving body and its environment receiving the dissipated energy. This composite system has the conservative Hamiltonian $H=K_1+V_1+H_2$ where $K_1$ is the kinetic energy of the moving body, $V_1$ its potential energy and $H_2$ the energy of the environment. The Lagrangian can be derived by using the usual Legendre transformation $L=2K_1+2K_2-H$ where $K_2$ is the total kinetic energy of the environment. An equivalent expression of this Lagrangian is $L=K_1-V_1-E_d$ where $E_d$ is the energy dissipated by the friction from the moving body into the environment from the beginning of the motion. The usual variation calculus of least action leads to the correct equation of the damped motion. We also show that this general formulation is a natural consequence of the virtual work principle.

cond-mat.stat-mech↗

arXiv2014

Learning directed acyclic graphs via bootstrap aggregating

Probabilistic graphical models are graphical representations of probability distributions. Graphical models have applications in many fields including biology, social sciences, linguistic, neuroscience. In this paper, we propose directed acyclic graphs (DAGs) learning via bootstrap aggregating. The proposed procedure is named as DAGBag. Specifically, an ensemble of DAGs is first learned based on bootstrap resamples of the data and then an aggregated DAG is derived by minimizing the overall distance to the entire ensemble. A family of metrics based on the structural hamming distance is defined for the space of DAGs (of a given node set) and is used for aggregation. Under the high-dimensional-low-sample size setting, the graph learned on one data set often has excessive number of false positive edges due to over-fitting of the noise. Aggregation overcomes over-fitting through variance reduction and thus greatly reduces false positives. We also develop an efficient implementation of the hill climbing search algorithm of DAG learning which makes the proposed method computationally competitive for the high-dimensional regime. The DAGBag procedure is implemented in the R package dagbag.

stat.ML↗

arXiv2013

Numerical study of path probability for stochastic motion of non dissipative systems

The path probability of stochastic motion of non dissipative or quasi-Hamiltonian systems is investigated by numerical experiment. The simulation model generates ideal one-dimensional motion of particles subject only to conservative forces in addition to Gaussian distributed random displacements. In the presence of dissipative forces, application of this ideal model requires that the dissipated energy is small with respect to the variation (work) of the conservative forces. The sample paths are sufficiently smooth space-time tubes with suitable width allowing correct evaluation of position, velocity, energy and action of each tube. It is found that the path probability decays exponentially with increasing action of the sample paths. The decay rate increases with decreasing Gaussian randomness.

cond-mat.stat-mech↗

arXiv2012

A dual modelling of evolving political opinion networks

We present the result of a dual modeling of opinion network. The model complements the agent-based opinion models by attaching to the social agent (voters) network a political opinion (party) network having its own intrinsic mechanisms of evolution. These two sub-networks form a global network which can be either isolated from or dependent on the external influence. Basically, the evolution of the agent network includes link adding and deleting, the opinion changes influenced by social validation, the political climate, the attractivity of the parties and the interaction between them. The opinion network is initially composed of numerous nodes representing opinions or parties which are located on a one dimensional axis according to their political positions. The mechanism of evolution includes union, splitting, change of position and of attractivity, taken into account the pairwise node interaction decaying with node distance in power law. The global evolution ends in a stable distribution of the social agents over a quasi-stable and fluctuating stationary number of remaining parties. Empirical study on the lifetime distribution of numerous parties and vote results is carried out to verify numerical results.

physics.soc-ph↗

arXiv2011

A note on logistic regression and logistic kernel machine models

This is a note on logistic regression models and logistic kernel machine models. It contains derivations to some of the expressions in a paper -- SNP Set Analysis for Detecting Disease Association Using Exon Sequence Data -- submitted to BMC proceedings by these authors.

stat.AP↗