112 pubs · 1,242 cites · h-index 19
Review highlights the role of somatic mutations in Alzheimer’s disease, suggesting novel therapeutic targets.
Path planning is a key problem in the autonomous navigation of mobile robots and a research hotspot in the field of robotics. Harris Hawk Optimization (HHO) faces challenges such as low solution accuracy and a slow convergence speed, and it easy falls into local optimization in path planning applications. For this reason, this paper proposes a Multi-strategy Improved Harris Hawk Optimization (MIHHO) algorithm. First, the double adaptive weight strategy is used to enhance the search capability of the algorithm to significantly improve the convergence accuracy and speed of path planning; second, the Dimension Learning-based Hunting (DLH) search strategy is introduced to effectively balance exploration and exploitation while maintaining the diversity of the population; and then, Position update strategy based on Dung Beetle Optimizer algorithm is proposed to reduce the algorithm's possibility of falling into local optimal solutions during path planning. The experimental results of the comparison of the test functions show that the MIHHO algorithm is ranked first in terms of performance, with significant improvements in optimization seeking ability, convergence speed, and stability. Finally, MIHHO is applied to robot path planning, and the test results show that in four environments with different complexities and scales, the average path lengths of MIHHO are improved by 1.99%, 14.45%, 4.52%, and 9.19% compared to HHO, respectively. These results indicate that MIHHO has significant performance advantages in path planning tasks and helps to improve the path planning efficiency and accuracy of mobile robots.
In the social context of the “Educational Involution”, the educational expectations of parents have a potential influence on the development of children. High parental educational expectations create parental anxiety, which in turn results in a rise in parental involvement and eventually promotes the growth and progress of children. The current study administered an electronic questionnaire to 891 parents of young children in four provinces of China. The questionnaire included the Parental Educational Expectations Scale, the Educational Anxiety Scale, the Parental Involvement Scale, and the Child Development Scale. This study used SPSS 27.0 for statistical data analysis and the SPSS macro program PROCESS to explore the mediation role. We found that (1) educational anxiety plays a mediating role between parental educational expectations and child development; (2) parental involvement has a mediating effect between parental educational expectations and child development; and (3) educational anxiety and parental involvement play a chain mediating role between parental educational expectations and child development. In conclusion, parental educational expectations appear to contribute to child development, and this effect may be mediated individually and sequentially by educational anxiety and parental involvement.
In many applications such as medical imaging, the measurement data represent counts of photons hitting a detector. Such counts in low-photon settings are often modeled using a Poisson distribution. However, this model assumes that the mean and variance of the signal's noise distribution are equal. For overdispersed data where the variance is greater than the mean, the negative binomial distribution is a more appropriate statistical model. In this paper, we propose an optimization approach for recovering images corrupted by overdispersed Poisson noise. In particular, we incorporate a weighted anisotropic-isotropic total variation regularizer, which avoids staircasing artifacts that are introduced by a regular total variation penalty. We use an alternating direction method of multipliers, where each subproblem has a closed-form solution. Numerical experiments demonstrate the effectiveness of our proposed approach, especially in very photon-limited settings.
The pretraining-finetuning paradigm has gained popularity in various computer vision tasks. In this paradigm, the emergence of active finetuning arises due to the abundance of large-scale data and costly annotation requirements. Active finetuning involves selecting a subset of data from an unlabeled pool for annotation, facilitating subsequent finetuning. However, the use of a limited number of training samples can lead to a biased distribution, potentially resulting in model overfitting. In this paper, we propose a new method called ActiveDC for the active finetuning tasks. Firstly, we select samples for annotation by optimizing the distribution similarity between the subset to be selected and the entire unlabeled pool in continuous space. Secondly, we calibrate the distribution of the selected samples by exploiting implicit category information in the unlabeled pool. The feature visualization provides an intuitive sense of the effectiveness of our method to distribution calibration. We conducted extensive experiments on three image classification datasets with different sampling ratios. The results indicate that ActiveDC consistently outperforms the baseline performance in all image classification tasks. The improvement is particularly significant when the sampling ratio is low, with performance gains of up to 10%. Our code will be publicly available.
The negative binomial model, which generalizes the Poisson distribution model, can be found in applications involving low-photon signal recovery, including medical imaging. Recent studies have explored several regularization terms for the negative binomial model, such as the ₚ quasi-norm with $0 < p < 1$, ₁ norm, and the total variation (TV) quasi-seminorm for promoting sparsity in signal recovery. These penalty terms have been shown to improve image reconstruction outcomes. In this paper, we investigate the ₚ quasi-seminorm, both isotropic and anisotropic ₚ TV quasi-seminorms, within the framework of the negative binomial statistical model. This problem can be formulated as an optimization problem, which we solve using a gradient-based approach. We present comparisons between the negative binomial and Poisson statistical models using the ₚ TV quasi-seminorm as well as common penalty terms. Our experimental results highlight the efficacy of the proposed method.
With the growing humanlike nature of dialog agents, people are now engaging in extended conversations that can stretch from brief moments to substantial periods of time. Understanding the factors that contribute to sustaining these interactions is crucial, yet existing studies primarily focusing on short-term simulations that rarely explore such prolonged and real conversations. In this paper, we investigate the factors influencing retention rates in real interactions with roleplaying models. By analyzing a large dataset of interactions between real users and thousands of characters, we systematically examine multiple factors and assess their impact on user retention rate. Surprisingly, we find that the degree to which the bot embodies the roles it plays has limited influence on retention rates, while the length of each turn it speaks significantly affects retention rates. This study sheds light on the critical aspects of user engagement with role-playing models and provides valuable insights for future improvements in the development of large language models for role-playing purposes.
Cancer immunotherapy, exemplified by the remarkable clinical benefits of the immune checkpoint blockade and chimeric antigen receptor T-cell therapy, is revolutionizing cancer therapy. They induce long-term tumor regression and overall survival benefit in many types of cancer. With the advances in our knowledge about the tumor immune microenvironment, remarkable progress has been made in the development of small-molecule drugs for immunotherapy. Small molecules targeting PRR-associated pathways, immune checkpoints, oncogenic signaling, metabolic pathways, cytokine/chemokine signaling, and immune-related kinases have been extensively investigated. Monotherapy of small-molecule immunotherapeutic drugs and their combinations with other antitumor modalities are under active clinical investigations to overcome immune tolerance and circumvent immune checkpoint inhibitor resistance. Here, we review the latest development of small-molecule agents for cancer immunotherapy by targeting defined pathways and highlighting their progress in recent clinical investigations.
Using numerical simulation methods in additive manufacturing (AM) processes can effectively reduce the cost of process exploration.This study presents a multi-scale simulation study based on the powder bed model for grain growth in the laser powder bed fusion (L-PBF) process of the Inconel 718 material.The finite volume method (FVM) and cellular automaton (CA) model were used to reproduce the process.The simulation results were consistent with the experimental results in terms of melt pool morphology, grain structure, and average grain size.The processes of the competitive and epitaxial growth of grains were discussed.The effects of laser power, scanning speed, hatch distance, and scanning strategy on microstructure were further analyzed through pole figure, grain area distribution, CA nucleation and capture event.The results demonstrated the feasibility of this method in predicting the evolution of microstructures in the L-PBF process and provide a reference for future research.
Nowadays, various multimodal fusion-based 3-D object detectors appear to provide a potential opportunity to solve the failure cases in single-modality methods. However, current fusion approaches still face some unfavorable factors, e.g., poor illumination driving conditions and crowded traffic scenarios, which will cause unsatisfying image quality and objects’ occlusion. To this end, we present a multimodal fusion network FSFNet consisting of local graph-aware point cloud backbone (LGB), foreground score-aware fusion network (FSFN), and the proposals’ refining loss (PRL) for the 3-D object detection task in this article. Concretely, the directed graph is built to generate edgewise features for each point, and the point features are supplemented with graph information in LGB. To alleviate the defect of undesirable image quality features caused by poor illumination condition, FSFN is designed to produce an adaptive multimodal feature by taking pointwise foreground scores into consideration. Hence, levelwise point features with high confidence are fully used, and the imperfect image information is suppressed in fusion stage. We further introduce PRL to reduce the false positive and false negative cases in crowded scenes by optimizing the location and scores of predicted 3-D bounding boxes. Extensive experiments conducted on the KITTI benchmark demonstrate that FSFNet owns its superiority over state-of-the-art networks. Moreover, FSFN is verified to be robust against the image inputs under poor illumination conditions.
Recently vision transformer models have become prominent models for a range of tasks. These models, however, usually suffer from intensive computational costs and heavy memory requirements, making them impractical for deployment on edge platforms. Recent studies have proposed to prune transformers in an unexplainable manner, which overlook the relationship between internal units of the model and the target class, thereby leading to inferior performance. To alleviate this problem, we propose a novel explainable pruning framework dubbed X-Pruner, which is designed by considering the explainability of the pruning criterion. Specifically, to measure each prunable unit's contribution to predicting each target class, a novel explainability-aware mask is proposed and learned in an end-to-end manner. Then, to preserve the most informative units and learn the layer-wise pruning rate, we adaptively search the layer-wise threshold that differentiates between unpruned and pruned units based on their explainability-aware mask values. To verify and evaluate our method, we apply the X-Pruner on representative transformer models including the DeiT and Swin Transformer. Comprehensive simulation results demonstrate that the proposed X-Pruner outperforms the state-of-the-art black-box methods with significantly reduced computational costs and slight performance degradation. Code is available at https://github.com/vickyyu90/XPruner.
Continual learning aims to train a model incrementally on a sequence of tasks without forgetting previous knowledge. Although continual learning has been widely studied in computer vision, its application to Vision+Language tasks is not that straightforward, as settings can be parameterized in multiple ways according to their input modalities. In this paper, we present a detailed study of how different settings affect performance for Visual Question Answering. We first propose three plausible task formulations and demonstrate their impact on the performance of continual learning algorithms. We break down several factors of task similarity, showing that performance and sensitivity to task order highly depend on the shift of the output distribution. We also investigate the potential of pretrained models and compare the robustness of transformer models with different visual embeddings. Finally, we provide an analysis interpreting model representations and their impact on forgetting. Our results highlight the importance of stabilizing visual representations in deeper layers.
Recent self-supervised learning methods are able to learn high-quality image representations and are closing the gap with supervised approaches. However, these methods are unable to acquire new knowledge incrementally – they are, in fact, mostly used only as a pre-training phase over IID data. In this work we investigate self-supervised methods in continual learning regimes without any replay mechanism. We show that naive functional regularization, also known as feature distillation, leads to lower plasticity and limits continual learning performance. Instead, we propose Projected Functional Regularization in which a separate temporal projection network ensures that the newly learned feature space preserves information of the previous one, while at the same time allowing for the learning of new features. This prevents forgetting while maintaining the plasticity of the learner. Comparison with other incremental learning approaches applied to self-supervision demonstrates that our method obtains competitive performance in different scenarios and on multiple datasets.
Attention mechanism has been ubiquitous in neural machine translation by dynamically selecting relevant contexts for different translations. Apart from performance gains, attention weights assigned to input tokens are often utilized to explain that high-attention tokens contribute more to the prediction. However, many works question whether this assumption holds in text classification by manually manipulating attention weights and observing decision flips. This article extends this question to Transformer-based neural machine translation, which heavily relies on cross-lingual attention to produce accurate translations but is relatively understudied in this context. We first design a mask perturbation model which automatically assesses each input’s contribution to model outputs. We then test whether the token contributing most to the current translation receives the highest attention weight. We find that it sometimes does not, which closely depends on the entropy of attention weights, the syntactic role of the current generation, and language pairs. We also rethink the discrepancy between attention weights and word alignments from the view of unreliable attention weights. Our observations further motivate us to calibrate the cross-lingual multi-head attention by attaching more attention to indispensable tokens, whose removal leads to a dramatic performance drop. Empirical experiments on different-scale translation tasks and text summarization tasks demonstrate that our calibration methods significantly outperform strong baselines.
Cross-lingual entity alignment identifies entity pairs that share the same meanings but locate in different language knowledge graphs (KGs). The study in this paper is to address two limitations that widely exist in current solutions: 1) the alignment loss functions defined at the entity level serve well the purpose of aligning labeled entities but fail to match the whole picture of labeled and unlabeled entities in different KGs; 2) the translation from one domain to the other has been considered (e.g., X to Y by M1 or Y to X by M2). However, the important duality of alignment between different KGs (X to Y by M1 and Y to X by M2) is ignored. We propose a novel entity alignment framework (OTEA), which dually optimizes the entity-level loss and group-level loss via optimal transport theory. We also impose a regularizer on the dual translation matrices to mitigate the effect of noise during transformation. Extensive experimental results show that our model consistently outperforms the state-of-the-arts with significant improvements on alignment accuracy.
Entity alignment associates entities in different knowledge graphs if they are semantically same, and has been successfully used in the knowledge graph construction and connection. Most of the recent solutions for entity alignment are based on knowledge graph embedding, which maps knowledge entities in a low-dimension space where entities are connected with the guidance of prior aligned entity pairs. The study in this paper focuses on two important issues that limit the accuracy of current entity alignment solutions: 1) labeled data of priorly aligned entity pairs are difficult and expensive to acquire, whereas abundant of unlabeled data are not used; and 2) knowledge graph embedding is affected by entity's degree difference, which brings challenges to align high frequent and low frequent entities. We propose a semi-supervised entity alignment method (SEA) to leverage both labeled entities and the abundant unlabeled entity information for the alignment. Furthermore, we improve the knowledge graph embedding with awareness of the degree difference by performing the adversarial training. To evaluate our proposed model, we conduct extensive experiments on real-world datasets. The experimental results show that our model consistently outperforms the state-of-the-art methods with significant improvement on alignment accuracy.
Histone deacetylation is one of the well characterized post-translational modifications related to transcriptional repression in eukaryotes. The process of histone deacetylation is achieved by histone deacetylases (HDACs). Over the last decade, substantial advances in our understanding of the mechanism of fruit ripening have been achieved, but the role of HDACs in this process has not been elucidated. In our study, an RNA interference (RNAi) expression vector targeting SlHDA1 was constructed and transformed into tomato plants. Shorter fruit ripening time and decreased storability were observed in SlHDA1 RNAi lines. The accumulation of carotenoid was increased through an alteration of the carotenoid pathway flux. Ethylene content, ethylene biosynthesis genes (ACS2, ACS4 and ACO1, ACO3) and ripening-associated genes (RIN, E4, E8, Cnr, TAGL1, PG, Pti4 and LOXB) were significantly up-regulated in SlHDA1 RNAi lines. In addition, the expression of fruit cell wall metabolism genes (HEX, MAN, TBG4, XTH5 and XYL) was enhanced compared with wild type. Furthermore, SlHDA1 RNAi seedlings displayed shorter hypocotyls and were more sensitive to ACC (1-aminocyclopropane-1-carboxylate) than the wild type. The results of our study indicate that SlHDA1 functions as a negative regulator of fruit ripening by affecting ethylene synthesis and carotenoid accumulation.
Considering the images captured under hazy weather conditions are blurred, a new dehazing algorithm based on overlapped sub-block homomorphic filtering in HSV color space is proposed. Firstly, the hazy image is transformed from RGB to HSV color space. Secondly, the luminance component V is dealt with the overlapped sub-block homomorphic filtering. Finally, the processed image is converted from HSV to RGB color space once again. Then the dehazing images will be obtained. According to the established algorithm model, the dehazing images could be evaluated by six objective evaluation parameters including average value, standard deviation, entropy, average gradient, edge intensity and contrast. The experimental results show that this algorithm has good dehazing effect. It can not only improve degradation of the image, but also amplify the image details and enhance the contrast of the image effectively.
Biclustering structures in data matrices were first formalized in a seminal paper by John Hartigan (1972) where one seeks to cluster cases and variables simultaneously. Such structures are also prevalent in block modeling of networks. In this paper, we develop a unified theory for the estimation and completion of matrices with biclustering structures, where the data is a partially observed and noise contaminated data matrix with a certain biclustering structure. In particular, we show that a constrained least squares estimator achieves minimax rate-optimal performance in several of the most important scenarios. To this end, we derive unified high probability upper bounds for all sub-Gaussian data and also provide matching minimax lower bounds in both Gaussian and binary cases. Due to the close connection of graphon to stochastic block models, an immediate consequence of our general results is a minimax rate-optimal estimator for sparse graphons.
Network analysis is becoming one of the most active research areas in statistics. Significant advances have been made recently on developing theories, methodologies and algorithms for analyzing networks. However, there has been little fundamental study on optimal estimation. In this paper, we establish optimal rate of convergence for graphon estimation. For the stochastic block model with k clusters, we show that the optimal rate under the mean squared error is n⁻¹log k+k²/n². The minimax upper bound improves the existing results in literature through a technique of solving a quadratic equation. When k≤√nlog n, as the number of the cluster k grows, the minimax rate grows slowly with only a logarithmic order n⁻¹log k. A key step to establish the lower bound is to construct a novel subset of the parameter space and then apply Fano’s lemma, from which we see a clear distinction of the nonparametric graphon estimation problem from classical nonparametric regression, due to the lack of identifiability of the order of nodes in exchangeable random graph models. As an immediate application, we consider nonparametric graphon estimation in a Hölder class with smoothness α. When the smoothness α≥1, the optimal rate of convergence is n⁻¹log n, independent of α, while for α∈(0,1), the rate is n-2α/(α+1), which is, to our surprise, identical to the classical nonparametric rate.