arXiv subjects

Khanh Nguyen

Publications and source records attributed to Khanh Nguyen.

At least 55 records · Page 3Linked to original sources

arXiv2022

Pressure and temperature dependence of fluorescence anisotropy of Green Fluorescent Protein

We have studied the effect of high hydrostatic pressure and temperature on the steady state fluorescence anisotropy of Green Fluorescent Protein (GFP). We find that the fluorescence anisotropy of GFP at a constant temperature decreases with increasing pressure. At atmospheric pressure, anisotropy decreases with increasing temperature but exhibits a maximum with temperature for pressure larger than 20 MPa. The temperature corresponding to the maximum of anisotropy increases with increasing pressure. By taking into account of the rotational correlation time changes of GFP with the pressure-temperature dependent viscosity of the solvent, we argue that viscosity increase with pressure is not a major contributing factor to the decrease in anisotropy with pressure. Our results suggest that the decrease of fluorescence anisotropy with pressure may result from changes in H-bonding environment around the chromophore.

q-bio.BM↗

arXiv2022

On Limits at Infinity of Weighted Sobolev Functions

We study necessary and sufficient conditions for a Muckenhoupt weight $w \in L^1_{\mathrm{loc}}(\mathbb R^d)$ that yield almost sure existence of radial, and vertical, limits at infinity for Sobolev functions $u \in W^{1,p}_{\mathrm{loc}}(\mathbb R^d,w)$ with a $p$-integrable gradient $|\nabla u|\in L^p(\mathbb R^d,w)$. The question is shown to subtly depend on the sense in which the limit is taken. First, we fully characterize the existence of radial limits. Second, we give essentially sharp sufficient conditions for the existence of vertical limits. In the specific setting of product and radial weights, we give if and only if statements. These generalize and give new proofs for results of Fefferman and Uspenski\uı.

math.AP↗

arXiv2021

Global-Local Attention for Emotion Recognition

Human emotion recognition is an active research area in artificial intelligence and has made substantial progress over the past few years. Many recent works mainly focus on facial regions to infer human affection, while the surrounding context information is not effectively utilized. In this paper, we proposed a new deep network to effectively recognize human emotions using a novel global-local attention mechanism. Our network is designed to extract features from both facial and context regions independently, then learn them together using the attention module. In this way, both the facial and contextual information is used to infer human emotions, therefore enhancing the discrimination of the classifier. The intensive experiments show that our method surpasses the current state-of-the-art methods on recent emotion datasets by a fair margin. Qualitatively, our global-local attention module can extract more meaningful attention maps than previous methods. The source code and trained model of our network are available at https://github.com/minhnhatvt/glamor-net

cs.CV↗

arXiv2021

Interactive Learning from Activity Description

We present a novel interactive learning protocol that enables training request-fulfilling agents by verbally describing their activities. Unlike imitation learning (IL), our protocol allows the teaching agent to provide feedback in a language that is most appropriate for them. Compared with reward in reinforcement learning (RL), the description feedback is richer and allows for improved sample complexity. We develop a probabilistic framework and an algorithm that practically implements our protocol. Empirical results in two challenging request-fulfilling problems demonstrate the strengths of our approach: compared with RL baselines, it is more sample-efficient; compared with IL baselines, it achieves competitive success rates without requiring the teaching agent to be able to demonstrate the desired behavior using the learning agent's actions. Apart from empirical evaluation, we also provide theoretical guarantees for our algorithm under certain assumptions about the teacher and the environment.

cs.CL↗

arXiv2020

Thermodynamically consistent nonlinear viscoplastic formulation with well-conditioned recovery of the inviscid solution: Theory and implicit integration algorithm with exact solution for the linear case

In this work, a consistent viscoplasticity formulation is derived from thermodynamical principles and employing the concept of continuum elastic corrector rate. The proposed model is developed based on the principle of maximum viscoplastic dissipation for determining the flow direction. The model uses both the equivalent viscoplastic strain and its rate as state variables. Power balance and energy balance give, respectively, separate evolution equations for the equivalent viscoplastic strain rate and the viscoplastic strain, the former written in terms of inviscid rates. Several key points distinguish our formulation from other proposals. First, the viscoplastic strain rate (instead of a yield function) consistently distinguishes conservative from dissipative behaviours during reverse loading; and the discrete implicit integration algorithm is an immediate implementation of the continuum theory based on the mentioned principles. Second, the inviscid solution is recovered in a well-conditioned manner by simply setting the viscosity to zero. Indeed, inviscid plasticity, viscoelasticity and viscoplasticity are particular cases of our formulation and integration algorithm, and are recovered just by setting the corresponding parameters to zero (viscosity or yield stress). Third, the linear viscoplasticity solution is obtained in an exact manner for proportional loading cases, independently of the time step employed. Four, general nonlinear models (Perzyna, Norton, etc) may be immediately incorporated as particular cases both in the theory and the computational implementation.

cond-mat.soft↗

arXiv2020

On the network orientational affinity assumption in polymers and the micro-macro connection through the chain stretch

We question the network affinity assumption in modeling chain orientations under polymer deformations, and the use of the stretch measure projected from the right Cauchy-Green deformation tensor (or non-affine micro-stretches derived from that measure) as a basic state variable for the micro-macro transition. These ingredients are standard, taken from the statistical theory of polymers, and used in most micromechanical polymer network and soft tissue models. The affinity assumption imposes a constraint in the network which results in an anisotropic distribution of the orientation of the chains and, hence, in an additional configurational entropy that should be included. This additional entropy would result in an additional stress tensor. But an arguably more natural alternative, in line with the typical assumption for the chain behavior itself and with the disregard of these forces, is to consider that the network may fluctuate unconstrained to adapt to macroscopic deformations. This way, the isotropic statistical distribution of the orientation of the chains is maintained unconstrained during deformation and no additional stress is imposed. Then, we show that this free-fluctuating network assumption is equivalent to consider the stretch projected from the stretch tensor (instead of the right Cauchy-Green deformation tensor) as the state variable for the deformation of the network chains. We show very important differences in predictions using both assumed behaviors, and demonstrate that with the free-fluctuating network assumption, we can obtain accurate predictions for all tests in polymers using just one test curve to calibrate the model. With the same macro-micro-macro approach employing the network affinity assumption, we are capable of capturing accurately only the test used for calibration of the model, but not the overall polymer behavior.

cond-mat.soft↗

arXiv2020

Global Voices: Crossing Borders in Automatic News Summarization

We construct Global Voices, a multilingual dataset for evaluating cross-lingual summarization methods. We extract social-network descriptions of Global Voices news articles to cheaply collect evaluation data for into-English and from-English summarization in 15 languages. Especially, for the into-English summarization task, we crowd-source a high-quality evaluation dataset based on guidelines that emphasize accuracy, coverage, and understandability. To ensure the quality of this dataset, we collect human ratings to filter out bad summaries, and conduct a survey on humans, which shows that the remaining summaries are preferred over the social-network summaries. We study the effect of translation quality in cross-lingual summarization, comparing a translate-then-summarize approach with several baselines. Our results highlight the limitations of the ROUGE metric that are overlooked in monolingual summarization. Our dataset is available for download at https://forms.gle/gpkJDT6RJWHM1Ztz9 .

cs.CL↗

arXiv2020

Active Imitation Learning from Multiple Non-Deterministic Teachers: Formulation, Challenges, and Algorithms

We formulate the problem of learning to imitate multiple, non-deterministic teachers with minimal interaction cost. Rather than learning a specific policy as in standard imitation learning, the goal in this problem is to learn a distribution over a policy space. We first present a general framework that efficiently models and estimates such a distribution by learning continuous representations of the teacher policies. Next, we develop Active Performance-Based Imitation Learning (APIL), an active learning algorithm for reducing the learner-teacher interaction cost in this framework. By making query decisions based on predictions of future progress, our algorithm avoids the pitfalls of traditional uncertainty-based approaches in the face of teacher behavioral uncertainty. Results on both toy and photo-realistic navigation tasks show that APIL significantly reduces the numbers of interactions with teachers without compromising on performance. Moreover, it is robust to various degrees of teacher behavioral uncertainty.

cs.LG↗

arXiv2020

Pre-processing Image using Brightening, CLAHE and RETINEX

This paper focuses on finding the most optimal pre-processing methods considering three common algorithms for image enhancement: Brightening, CLAHE and Retinex. For the purpose of image training in general, these methods will be combined to find out the most optimal method for image enhancement. We have carried out the research on the different permutation of three methods: Brightening, CLAHE and Retinex. The evaluation is based on Canny Edge detection applied to all processed images. Then the sharpness of objects will be justified by true positive pixels number in comparison between images. After using different number combinations pre-processing functions on images, CLAHE proves to be the most effective in edges improvement, Brightening does not show much effect on the edges enhancement, and the Retinex even reduces the sharpness of images and shows little contribution on images enhancement.

eess.IV↗

arXiv2019

Help, Anna! Visual Navigation with Natural Multimodal Assistance via Retrospective Curiosity-Encouraging Imitation Learning

Mobile agents that can leverage help from humans can potentially accomplish more complex tasks than they could entirely on their own. We develop "Help, Anna!" (HANNA), an interactive photo-realistic simulator in which an agent fulfills object-finding tasks by requesting and interpreting natural language-and-vision assistance. An agent solving tasks in a HANNA environment can leverage simulated human assistants, called ANNA (Automatic Natural Navigation Assistants), which, upon request, provide natural language and visual instructions to direct the agent towards the goals. To address the HANNA problem, we develop a memory-augmented neural agent that hierarchically models multiple levels of decision-making, and an imitation learning algorithm that teaches the agent to avoid repeating past mistakes while simultaneously predicting its own chances of making future progress. Empirically, our approach is able to ask for help more effectively than competitive baselines and, thus, attains higher task success rate on both previously seen and previously unseen environments. We publicly release code and data at https://github.com/khanhptnk/hanna . A video demo is available at https://youtu.be/18P94aaaLKg .

cs.HC↗

arXiv2019

On Deep Domain Adaptation: Some Theoretical Understandings

Compared with shallow domain adaptation, recent progress in deep domain adaptation has shown that it can achieve higher predictive performance and stronger capacity to tackle structural data (e.g., image and sequential data). The underlying idea of deep domain adaptation is to bridge the gap between source and target domains in a joint space so that a supervised classifier trained on labeled source data can be nicely transferred to the target domain. This idea is certainly intuitive and powerful, however, limited theoretical understandings have been developed to support its underpinning principle. In this paper, we have provided a rigorous framework to explain why it is possible to close the gap of the target and source domains in the joint space. More specifically, we first study the loss incurred when performing transfer learning from the source to the target domain. This provides a theory that explains and generalizes existing work in deep domain adaptation which was mainly empirical. This enables us to further explain why closing the gap in the joint space can directly minimize the loss incurred for transfer learning between the two domains. To our knowledge, this offers the first theoretical result that characterizes a direct bound on the joint space and the gain of transfer learning via deep domain adaptation

cs.LG↗

arXiv2019

Vision-based Navigation with Language-based Assistance via Imitation Learning with Indirect Intervention

We present Vision-based Navigation with Language-based Assistance (VNLA), a grounded vision-language task where an agent with visual perception is guided via language to find objects in photorealistic indoor environments. The task emulates a real-world scenario in that (a) the requester may not know how to navigate to the target objects and thus makes requests by only specifying high-level end-goals, and (b) the agent is capable of sensing when it is lost and querying an advisor, who is more qualified at the task, to obtain language subgoals to make progress. To model language-based assistance, we develop a general framework termed Imitation Learning with Indirect Intervention (I3L), and propose a solution that is effective on the VNLA task. Empirical results show that this approach significantly improves the success rate of the learning agent over other baselines in both seen and unseen environments. Our code and data are publicly available at https://github.com/debadeepta/vnla .

cs.LG↗

arXiv2017

Reinforcement Learning for Bandit Neural Machine Translation with Simulated Human Feedback

Machine translation is a natural candidate problem for reinforcement learning from human feedback: users provide quick, dirty ratings on candidate translations to guide a system to improve. Yet, current neural machine translation training focuses on expensive human-generated reference translations. We describe a reinforcement learning algorithm that improves neural machine translation systems from simulated human feedback. Our algorithm combines the advantage actor-critic algorithm (Mnih et al., 2016) with the attention-based neural encoder-decoder architecture (Luong et al., 2015). This algorithm (a) is well-designed for problems with a large action space and delayed rewards, (b) effectively optimizes traditional corpus-level machine translation metrics, and (c) is robust to skewed, high-variance, granular feedback modeled after actual human behaviors.

cs.CL↗

arXiv2017

Analogical-based Bayesian Optimization

Some real-world problems revolve to solve the optimization problem \max_{x\in\mathcal{X}}f\left(x\right) where f\left(.\right) is a black-box function and X might be the set of non-vectorial objects (e.g., distributions) where we can only define a symmetric and non-negative similarity score on it. This setting requires a novel view for the standard framework of Bayesian Optimization that generalizes the core insightful spirit of this framework. With this spirit, in this paper, we propose Analogical-based Bayesian Optimization that can maximize black-box function over a domain where only a similarity score can be defined. Our pathway is as follows: we first base on the geometric view of Gaussian Processes (GP) to define the concept of influence level that allows us to analytically represent predictive means and variances of GP posteriors and base on that view to enable replacing kernel similarity by a more genetic similarity score. Furthermore, we also propose two strategies to find a batch of query points that can efficiently handle high dimensional data.

cs.LG↗

arXiv2017

The UMD Neural Machine Translation Systems at WMT17 Bandit Learning Task

We describe the University of Maryland machine translation systems submitted to the WMT17 German-English Bandit Learning Task. The task is to adapt a translation system to a new domain, using only bandit feedback: the system receives a German sentence to translate, produces an English sentence, and only gets a scalar score as feedback. Targeting these two challenges (adaptation and bandit learning), we built a standard neural machine translation system and extended it in two ways: (1) robust reinforcement learning techniques to learn effectively from the bandit feedback, and (2) domain adaptation using data selection from a large corpus of parallel data.

cs.CL↗

arXiv2017

Scalable Semi-supervised Learning with Graph-based Kernel Machine

Acquiring labels are often costly, whereas unlabeled data are usually easy to obtain in modern machine learning applications. Semi-supervised learning provides a principled machine learning framework to address such situations, and has been applied successfully in many real-word applications and industries. Nonetheless, most of existing semi-supervised learning methods encounter two serious limitations when applied to modern and large-scale datasets: computational burden and memory usage demand. To this end, we present in this paper the Graph-based semi-supervised Kernel Machine (GKM), a method that leverages the generalization ability of kernel-based method with the geometrical and distributive information formulated through a spectral graph induced from data for semi-supervised learning purpose. Our proposed GKM can be solved directly in the primal form using the Stochastic Gradient Descent method with the ideal convergence rate $O(\frac{1}{T})$. Besides, our formulation is suitable for a wide spectrum of important loss functions in the literature of machine learning (e.g., Hinge, smooth Hinge, Logistic, L1, and ε-insensitive) and smoothness functions (i.e., $l_p(t) = |t|^p$ with $p\ge1$). We further show that the well-known Laplacian Support Vector Machine is a special case of our formulation. We validate our proposed method on several benchmark datasets to demonstrate that GKM is appropriate for the large-scale datasets since it is optimal in memory usage and yields superior classification accuracy whilst simultaneously achieving a significant computation speed-up in comparison with the state-of-the-art baselines.

cs.LG↗

arXiv2017

Morphology, cell division, and viability of Saccharomyces cerevisiae at high hydrostatic pressure

High hydrostatic pressure is commonly encountered in many environments, but the effects of high pressure on eukaryotic cells have been understudied. To understand the effects of hydrostatic pressure in the model eukaryote, Saccharomyces cerevisiae, we have performed quantitative experiments of cell division, cell morphology, and cell death under a wide range of pressures. We developed an automated image analysis method for quantification of the yeast budding index - a measure of cell cycle state - as well as a continuum model of budding to investigate the effect of pressure on cell division and cell morphology. We find that the budding index, the average cell size, and the eccentricity - a measure of how much the cell morphology varies from the being elliptical - of the cells decrease with increasing pressure. Furthermore, high hydrostatic pressure led to the small but finite probability of cell death via both apoptosis and necrosis. Our experiments suggest that decrease of budding index arises from cellular arrest or death at the cell cycle checkpoints during different stages of cell division.

q-bio.CB↗

arXiv2016

Imitation Learning with Recurrent Neural Networks

We present a novel view that unifies two frameworks that aim to solve sequential prediction problems: learning to search (L2S) and recurrent neural networks (RNN). We point out equivalences between elements of the two frameworks. By complementing what is missing from one framework comparing to the other, we introduce a more advanced imitation learning framework that, on one hand, augments L2S s notion of search space and, on the other hand, enhances RNNs training procedure to be more robust to compounding errors arising from training on highly correlated examples.

cs.CL↗