arXiv subjects

Xiao Zhou

Publications and source records attributed to Xiao Zhou.

At least 127 records · Page 7Linked to original sources

arXiv2023

Black-box Prompt Learning for Pre-trained Language Models

The increasing scale of general-purpose Pre-trained Language Models (PLMs) necessitates the study of more efficient adaptation across different downstream tasks. In this paper, we establish a Black-box Discrete Prompt Learning (BDPL) to resonate with pragmatic interactions between the cloud infrastructure and edge devices. Particularly, instead of fine-tuning the model in the cloud, we adapt PLMs by prompt learning, which efficiently optimizes only a few parameters of the discrete prompts. Moreover, we consider the scenario that we do not have access to the parameters and gradients of the pre-trained models, except for its outputs given inputs. This black-box setting secures the cloud infrastructure from potential attack and misuse to cause a single-point failure, which is preferable to the white-box counterpart by current infrastructures. Under this black-box constraint, we apply a variance-reduced policy gradient algorithm to estimate the gradients of parameters in the categorical distribution of each discrete prompt. In light of our method, the user devices can efficiently tune their tasks by querying the PLMs bounded by a range of API calls. Our experiments on RoBERTa and GPT-3 demonstrate that the proposed algorithm achieves significant improvement on eight benchmarks in a cloud-device collaboration manner. Finally, we conduct in-depth case studies to comprehensively analyze our method in terms of various data sizes, prompt lengths, training budgets, optimization objectives, prompt transferability, and explanations of the learned prompts. Our code will be available at https://github.com/shizhediao/Black-Box-Prompt-Learning.

cs.CL↗

arXiv2023

Model Agnostic Sample Reweighting for Out-of-Distribution Learning

Distributionally robust optimization (DRO) and invariant risk minimization (IRM) are two popular methods proposed to improve out-of-distribution (OOD) generalization performance of machine learning models. While effective for small models, it has been observed that these methods can be vulnerable to overfitting with large overparameterized models. This work proposes a principled method, \textbf{M}odel \textbf{A}gnostic sam\textbf{PL}e r\textbf{E}weighting (\textbf{MAPLE}), to effectively address OOD problem, especially in overparameterized scenarios. Our key idea is to find an effective reweighting of the training samples so that the standard empirical risk minimization training of a large model on the weighted training data leads to superior OOD generalization performance. The overfitting issue is addressed by considering a bilevel formulation to search for the sample reweighting, in which the generalization complexity depends on the search space of sample weights instead of the model size. We present theoretical analysis in linear case to prove the insensitivity of MAPLE to model size, and empirically verify its superiority in surpassing state-of-the-art methods by a large margin. Code is available at \url{https://github.com/x-zho14/MAPLE}.

cs.LG↗

arXiv2023

Probabilistic Bilevel Coreset Selection

The goal of coreset selection in supervised learning is to produce a weighted subset of data, so that training only on the subset achieves similar performance as training on the entire dataset. Existing methods achieved promising results in resource-constrained scenarios such as continual learning and streaming. However, most of the existing algorithms are limited to traditional machine learning models. A few algorithms that can handle large models adopt greedy search approaches due to the difficulty in solving the discrete subset selection problem, which is computationally costly when coreset becomes larger and often produces suboptimal results. In this work, for the first time we propose a continuous probabilistic bilevel formulation of coreset selection by learning a probablistic weight for each training sample. The overall objective is posed as a bilevel optimization problem, where 1) the inner loop samples coresets and train the model to convergence and 2) the outer loop updates the sample probability progressively according to the model's performance. Importantly, we develop an efficient solver to the bilevel optimization problem via unbiased policy gradient without trouble of implicit differentiation. We provide the convergence property of our training procedure and demonstrate the superiority of our algorithm against various coreset selection methods in various tasks, especially in more challenging label-noise and class-imbalance scenarios.

cs.LG↗

arXiv2022

GWRBoost:A geographically weighted gradient boosting method for explainable quantification of spatially-varying relationships

The geographically weighted regression (GWR) is an essential tool for estimating the spatial variation of relationships between dependent and independent variables in geographical contexts. However, GWR suffers from the problem that classical linear regressions, which compose the GWR model, are more prone to be underfitting, especially for significant volume and complex nonlinear data, causing inferior comparative performance. Nevertheless, some advanced models, such as the decision tree and the support vector machine, can learn features from complex data more effectively while they cannot provide explainable quantification for the spatial variation of localized relationships. To address the above issues, we propose a geographically gradient boosting weighted regression model, GWRBoost, that applies the localized additive model and gradient boosting optimization method to alleviate underfitting problems and retains explainable quantification capability for spatially-varying relationships between geographically located variables. Furthermore, we formulate the computation method of the Akaike information score for the proposed model to conduct the comparative analysis with the classic GWR algorithm. Simulation experiments and the empirical case study are applied to prove the efficient performance and practical value of GWRBoost. The results show that our proposed model can reduce the RMSE by 18.3% in parameter estimation accuracy and AICc by 67.3% in the goodness of fit.

cs.LG↗

arXiv2022

DouFu: A Double Fusion Joint Learning Method For Driving Trajectory Representation

Driving trajectory representation learning is of great significance for various location-based services, such as driving pattern mining and route recommendation. However, previous representation generation approaches tend to rarely address three challenges: 1) how to represent the intricate semantic intentions of mobility inexpensively; 2) complex and weak spatial-temporal dependencies due to the sparsity and heterogeneity of the trajectory data; 3) route selection preferences and their correlation to driving behavior. In this paper, we propose a novel multimodal fusion model, DouFu, for trajectory representation joint learning, which applies multimodal learning and attention fusion module to capture the internal characteristics of trajectories. We first design movement, route, and global features generated from the trajectory data and urban functional zones and then analyze them respectively with the attention encoder or feed forward network. The attention fusion module incorporates route features with movement features to create a better spatial-temporal embedding. With the global semantic feature, DouFu produces a comprehensive embedding for each trajectory. We evaluate representations generated by our method and other baseline models on classification and clustering tasks. Empirical results show that DouFu outperforms other models in most of the learning algorithms like the linear regression and the support vector machine by more than 10%.

cs.LG↗

arXiv2022

PUTN: A Plane-fitting based Uneven Terrain Navigation Framework

Autonomous navigation of ground robots has been widely used in indoor structured 2D environments, but there are still many challenges in outdoor 3D unstructured environments, especially in rough, uneven terrains. This paper proposed a plane-fitting based uneven terrain navigation framework (PUTN) to solve this problem. The implementation of PUTN is divided into three steps. First, based on Rapidly-exploring Random Trees (RRT), an improved sample-based algorithm called Plane Fitting RRT* (PF-RRT*) is proposed to obtain a sparse trajectory. Each sampling point corresponds to a custom traversability index and a fitted plane on the point cloud. These planes are connected in series to form a traversable strip. Second, Gaussian Process Regression is used to generate traversability of the dense trajectory interpolated from the sparse trajectory, and the sampling tree is used as the training set. Finally, local planning is performed using nonlinear model predictive control (NMPC). By adding the traversability index and uncertainty to the cost function, and adding obstacles generated by the real-time point cloud to the constraint function, a safe motion planning algorithm with smooth speed and strong robustness is available. Experiments in real scenarios are conducted to verify the effectiveness of the method. The source code is released for the reference of the community.

cs.RO↗

arXiv2022

Neural Global Shutter: Learn to Restore Video from a Rolling Shutter Camera with Global Reset Feature

Most computer vision systems assume distortion-free images as inputs. The widely used rolling-shutter (RS) image sensors, however, suffer from geometric distortion when the camera and object undergo motion during capture. Extensive researches have been conducted on correcting RS distortions. However, most of the existing work relies heavily on the prior assumptions of scenes or motions. Besides, the motion estimation steps are either oversimplified or computationally inefficient due to the heavy flow warping, limiting their applicability. In this paper, we investigate using rolling shutter with a global reset feature (RSGR) to restore clean global shutter (GS) videos. This feature enables us to turn the rectification problem into a deblur-like one, getting rid of inaccurate and costly explicit motion estimation. First, we build an optic system that captures paired RSGR/GS videos. Second, we develop a novel algorithm incorporating spatial and temporal designs to correct the spatial-varying RSGR distortion. Third, we demonstrate that existing image-to-image translation algorithms can recover clean GS videos from distorted RSGR inputs, yet our algorithm achieves the best performance with the specific designs. Our rendered results are not only visually appealing but also beneficial to downstream tasks. Compared to the state-of-the-art RS solution, our RSGR solution is superior in both effectiveness and efficiency. Considering it is easy to realize without changing the hardware, we believe our RSGR solution can potentially replace the RS solution in taking distortion-free videos with low noise and low budget.

cs.CV↗

arXiv2022

UNet#: A UNet-like Redesigning Skip Connections for Medical Image Segmentation

As an essential prerequisite for developing a medical intelligent assistant system, medical image segmentation has received extensive research and concentration from the neural network community. A series of UNet-like networks with encoder-decoder architecture has achieved extraordinary success, in which UNet2+ and UNet3+ redesign skip connections, respectively proposing dense skip connection and full-scale skip connection and dramatically improving compared with UNet in medical image segmentation. However, UNet2+ lacks sufficient information explored from the full scale, which will affect the learning of organs' location and boundary. Although UNet3+ can obtain the full-scale aggregation feature map, owing to the small number of neurons in the structure, it does not satisfy the segmentation of tiny objects when the number of samples is small. This paper proposes a novel network structure combining dense skip connections and full-scale skip connections, named UNet-sharp (UNet\#) for its shape similar to symbol \#. The proposed UNet\# can aggregate feature maps of different scales in the decoder sub-network and capture fine-grained details and coarse-grained semantics from the full scale, which benefits learning the exact location and accurately segmenting the boundary of organs or lesions. We perform deep supervision for model pruning to speed up testing and make it possible for the model to run on mobile devices; furthermore, designing two classification-guided modules to reduce false positives achieves more accurate segmentation results. Various experiments of semantic segmentation and instance segmentation on different modalities (EM, CT, MRI) and dimensions (2D, 3D) datasets, including the nuclei, brain tumor, liver, and lung, demonstrate that the proposed method outperforms state-of-the-art models.

eess.IV↗

arXiv2022

NTIRE 2022 Challenge on Super-Resolution and Quality Enhancement of Compressed Video: Dataset, Methods and Results

This paper reviews the NTIRE 2022 Challenge on Super-Resolution and Quality Enhancement of Compressed Video. In this challenge, we proposed the LDV 2.0 dataset, which includes the LDV dataset (240 videos) and 95 additional videos. This challenge includes three tracks. Track 1 aims at enhancing the videos compressed by HEVC at a fixed QP. Track 2 and Track 3 target both the super-resolution and quality enhancement of HEVC compressed video. They require x2 and x4 super-resolution, respectively. The three tracks totally attract more than 600 registrations. In the test phase, 8 teams, 8 teams and 12 teams submitted the final results to Tracks 1, 2 and 3, respectively. The proposed methods and solutions gauge the state-of-the-art of super-resolution and quality enhancement of compressed video. The proposed LDV 2.0 dataset is available at https://github.com/RenYang-home/LDV_dataset. The homepage of this challenge (including open-sourced codes) is at https://github.com/RenYang-home/NTIRE22_VEnh_SR.

cs.CV↗

arXiv2022

Effect of compositional fluctuation on the survival of bet-hedging species

Understanding the coexistence of diverse species in a changing environment is an important problem in community ecology. Bet-hedging is a strategy that helps species survive in such changing environments. However, studies of bet-hedging have often focused on the expected long-term growth rate of the species by itself, neglecting competition with other coexisting species. Here we study the extinction risk of a bet-hedging species in competition with others. We show that there are three contributions to the extinction risk. The first is the usual demographic fluctuation due to stochastic reproduction and selection processes in finite populations. The second, due to the fluctuation of population growth rate caused by environmental changes, may counterintuitively reduce the extinction risk for small populations. Besides those two, we reveal a third contribution, which is unique to bet-hedging species that diversify into multiple phenotypes: The phenotype composition of the population will fluctuate over time, resulting in increased extinction risk. We compare such compositional fluctuation to the demographic and environmental contributions, showing how they have different effects on the extinction risk depending on the population size, generation overlap, and environmental correlation.

physics.bio-ph↗

arXiv2022

Evaluation of non-pharmaceutical interventions and optimal strategies for containing the COVID-19 pandemic

Given multiple new COVID-19 variants are continuously emerging, non-pharmaceutical interventions are still primary control strategies to curb the further spread of coronavirus. However, implementing strict interventions over extended periods of time is inevitably hurting the economy. With an aim to solve this multi-objective decision-making problem, we investigate the underlying associations between policies, mobility patterns, and virus transmission. We further evaluate the relative performance of existing COVID-19 control measures and explore potential optimal strategies that can strike the right balance between public health and socio-economic recovery for individual states in the US. The results highlight the power of state of emergency declaration and wearing face masks and emphasize the necessity of pursuing tailor-made strategies for different states and phases of epidemiological transmission. Our framework enables policymakers to create more refined designs of COVID-19 strategies and can be extended to inform policy makers of any country about best practices in pandemic response.

q-bio.PE↗

arXiv2021

Efficient Neural Network Training via Forward and Backward Propagation Sparsification

Sparse training is a natural idea to accelerate the training speed of deep neural networks and save the memory usage, especially since large modern neural networks are significantly over-parameterized. However, most of the existing methods cannot achieve this goal in practice because the chain rule based gradient (w.r.t. structure parameters) estimators adopted by previous methods require dense computation at least in the backward propagation step. This paper solves this problem by proposing an efficient sparse training method with completely sparse forward and backward passes. We first formulate the training process as a continuous minimization problem under global sparsity constraint. We then separate the optimization process into two steps, corresponding to weight update and structure parameter update. For the former step, we use the conventional chain rule, which can be sparse via exploiting the sparse structure. For the latter step, instead of using the chain rule based gradient estimators as in existing methods, we propose a variance reduced policy gradient estimator, which only requires two forward passes without backward propagation, thus achieving completely sparse training. We prove that the variance of our gradient estimator is bounded. Extensive experimental results on real-world datasets demonstrate that compared to previous methods, our algorithm is much more effective in accelerating the training process, up to an order of magnitude faster.

cs.LG↗

arXiv2021

Effective Sparsification of Neural Networks with Global Sparsity Constraint

Weight pruning is an effective technique to reduce the model size and inference time for deep neural networks in real-world deployments. However, since magnitudes and relative importance of weights are very different for different layers of a neural network, existing methods rely on either manual tuning or handcrafted heuristic rules to find appropriate pruning rates individually for each layer. This approach generally leads to suboptimal performance. In this paper, by directly working on the probability space, we propose an effective network sparsification method called {\it probabilistic masking} (ProbMask), which solves a natural sparsification formulation under global sparsity constraint. The key idea is to use probability as a global criterion for all layers to measure the weight importance. An appealing feature of ProbMask is that the amounts of weight redundancy can be learned automatically via our constraint and thus we avoid the problem of tuning pruning rates individually for different layers in a network. Extensive experimental results on CIFAR-10/100 and ImageNet demonstrate that our method is highly effective, and can outperform previous state-of-the-art methods by a significant margin, especially in the high pruning rate situation. Notably, the gap of Top-1 accuracy between our ProbMask and existing methods can be up to 10\%. As a by-product, we show ProbMask is also highly effective in identifying supermasks, which are subnetworks with high performance in a randomly weighted dense neural network.

cs.LG↗

arXiv2021

Defect Engineered Room-Temperature Ferromagnetism in Quasi-Two-Dimensional Nitrided CoTa2O6

Thermal ammonolysis of quasi-two-dimensional (quasi-2D) CoTa2O6 yields the O2-/N3- and anionic vacancy ordered Co2+Ta5+2O6-xN2x/3$\Box$x/3 (x $\leq$ 0.15) that exhibits a transition from antiferromagnetism to defect engineered above room-temperature ferromagnetism as evidenced by diffraction, spectroscopic and magnetic characterizations. First-principles calculations reveal the origin of ferromagnetism is a particular CoON configuration with N located at Wyckoff position 8j, which breaks mirror symmetry about ab plane. A pressure-induced electronic phase transition is also predicted at around 24.5 GPa, accompanied by insulator-to-metal transition and magnetic moment vanishing.

cond-mat.mtrl-sci↗

arXiv2021

Subspace-based compressive sensing algorithm for raypath separation in a shallow-water waveguide

Compressive sensing (CS) has been applied to estimate the direction of arrival (DOA) in underwater acoustics. However, the key problem needed to be resolved in a {multipath} propagation environment is to suppress the interferences between the raypaths. Thus, in this paper, {a subspace-based compressive sensing algorithm that formulates the statistic information of the signal subspace in a CS framework is proposed.} The experiment results show that (1) the proposed algorithm enables the separation of raypaths that arrive closely at the {receiver} array and (2) the existing algorithms fail, especially in a low signal-to-noise ratio (SNR) environment.

cs.SD↗

arXiv2021

Tuning IR-cut Filter for Illumination-aware Spectral Reconstruction from RGB

To reconstruct spectral signals from multi-channel observations, in particular trichromatic RGBs, has recently emerged as a promising alternative to traditional scanning-based spectral imager. It has been proven that the reconstruction accuracy relies heavily on the spectral response of the RGB camera in use. To improve accuracy, data-driven algorithms have been proposed to retrieve the best response curves of existing RGB cameras, or even to design brand new three-channel response curves. Instead, this paper explores the filter-array based color imaging mechanism of existing RGB cameras, and proposes to design the IR-cut filter properly for improved spectral recovery, which stands out as an in-between solution with better trade-off between reconstruction accuracy and implementation complexity. We further propose a deep learning based spectral reconstruction method, which allows to recover the illumination spectrum as well. Experiment results with both synthetic and real images under daylight illumination have shown the benefits of our IR-cut filter tuning method and our illumination-aware spectral reconstruction method.

eess.IV↗

arXiv2020

Application of Seq2Seq Models on Code Correction

We apply various seq2seq models on programming language correction tasks on Juliet Test Suite for C/C++ and Java of Software Assurance Reference Datasets(SARD), and achieve 75\%(for C/C++) and 56\%(for Java) repair rates on these tasks. We introduce Pyramid Encoder in these seq2seq models, which largely increases the computational efficiency and memory efficiency, while remain similar repair rate to their non-pyramid counterparts. We successfully carry out error type classification task on ITC benchmark examples (with only 685 code instances) using transfer learning with models pre-trained on Juliet Test Suite, pointing out a novel way of processing small programing language datasets.

cs.SE↗

arXiv2020

Orbital period correction and light curve modeling of the W-subtype shallow contact binary OWLeo

Orbital period and multi-color light curves' investigation of OW Leo are presented for the first time. The orbital period of OW Leo is corrected from $P = 0.325545$ days to $P = 0.32554052$ days in our work, and the observational data from the All-Sky Automated Survey for SuperNovae (ASAS-SN) are used to test the newly determined orbital period. Then, the phased light curves are calculated with the new period and the Wilson-Devinney program is applied to model the light curves, which reveal that OW Leo is a W-subtype shallow contact binary system ($q = 3.05$, $f = 12.8\,\%$). The absolute physical parameters of the two component stars are estimated to be $M_{1} = 0.31(1)M_\odot$, $M_{2} = 0.95(3)M_\odot$, $R_{1} = 0.63(1)R_\odot$, $R_{2} = 1.04(1)R_\odot$, $L_{1} = 0.43(1)L_\odot$ and $L_{2} = 1.01(2)L_\odot$. The evolutionary status show that the more massive star is less evolved than the less massive star. OW Leo has very low metal abundance, which means its formation and evolution are hardly influenced by any additional component. It is formed from an initially detached binary systems through nuclear evolution and angular momentum loss via magnetic braking, and have passed a very long time of main sequence evolution.

astro-ph.SR↗