arXiv subjects

Yang Li

Publications and source records attributed to Yang Li.

At least 955 records · Page 53Linked to original sources

arXiv2023

PaLI-X: On Scaling up a Multilingual Vision and Language Model

We present the training recipe and results of scaling up PaLI-X, a multilingual vision and language model, both in terms of size of the components and the breadth of its training task mixture. Our model achieves new levels of performance on a wide-range of varied and complex tasks, including multiple image-based captioning and question-answering tasks, image-based document understanding and few-shot (in-context) learning, as well as object detection, video question answering, and video captioning. PaLI-X advances the state-of-the-art on most vision-and-language benchmarks considered (25+ of them). Finally, we observe emerging capabilities, such as complex counting and multilingual object detection, tasks that are not explicitly in the training mix.

cs.CV↗

arXiv2023

Improving Stability and Performance of Spiking Neural Networks through Enhancing Temporal Consistency

Spiking neural networks have gained significant attention due to their brain-like information processing capabilities. The use of surrogate gradients has made it possible to train spiking neural networks with backpropagation, leading to impressive performance in various tasks. However, spiking neural networks trained with backpropagation typically approximate actual labels using the average output, often necessitating a larger simulation timestep to enhance the network's performance. This delay constraint poses a challenge to the further advancement of SNNs. Current training algorithms tend to overlook the differences in output distribution at various timesteps. Particularly for neuromorphic datasets, inputs at different timesteps can cause inconsistencies in output distribution, leading to a significant deviation from the optimal direction when combining optimization directions from different moments. To tackle this issue, we have designed a method to enhance the temporal consistency of outputs at different timesteps. We have conducted experiments on static datasets such as CIFAR10, CIFAR100, and ImageNet. The results demonstrate that our algorithm can achieve comparable performance to other optimal SNN algorithms. Notably, our algorithm has achieved state-of-the-art performance on neuromorphic datasets DVS-CIFAR10 and N-Caltech101, and can achieve superior performance in the test phase with timestep T=1.

cs.NE↗

arXiv2023

First Principles and Machine Learning Identify Key Pairing Strength Factors of Cuprate Superconductors

By using band structure calculations of quantum mechanical theory, some important peaks of DoS (Density of States) were obtained and classified based on crystal structure laws of cuprate superconductivity. In particular, the orbital interactions of the in-plane and out-of-plane ions of the copper-oxygen plane were investigated. The position, half-width, and height of DOS peak features were collected for all 35 typical curate systems which have critical temperature maximum data from the works of literature. By training test of 7 common machine learning algorithms, the relationship between the Tc maximum values and these orbital interaction parameters were mined. It was found that the key features of the orbital interaction affecting the Tc maximum were not only the flat band but also a new interaction between core orbitals in a deeper energy band position.

cond-mat.supr-con↗

arXiv2023

Realistic Safety-critical Scenarios Search for Autonomous Driving System via Behavior Tree

The simulation-based testing of Autonomous Driving Systems (ADSs) has gained significant attention. However, current approaches often fall short of accurately assessing ADSs for two reasons: over-reliance on expert knowledge and the utilization of simplistic evaluation metrics. That leads to discrepancies between simulated scenarios and naturalistic driving environments. To address this, we propose the Matrix-Fuzzer, a behavior tree-based testing framework, to automatically generate realistic safety-critical test scenarios. Our approach involves the $log2BT$ method, which abstracts logged road-users' trajectories to behavior sequences. Furthermore, we vary the properties of behaviors from real-world driving distributions and then use an adaptive algorithm to explore the input space. Meanwhile, we design a general evaluation engine that guides the algorithm toward critical areas, thus reducing the generation of invalid scenarios. Our approach is demonstrated in our Matrix Simulator. The experimental results show that: (1) Our $log2BT$ achieves satisfactory trajectory reconstructions. (2) Our approach is able to find the most types of safety-critical scenarios, but only generating around 30% of the total scenarios compared with the baseline algorithm. Specifically, it improves the ratio of the critical violations to total scenarios and the ratio of the types to total scenarios by at least 10x and 5x, respectively, while reducing the ratio of the invalid scenarios to total scenarios by at least 58% in two case studies.

cs.RO↗

arXiv2023

Feature Chirality in Deep Learning Models

As deep learning applications extensively increase by leaps and bounds, their interpretability has become increasingly prominent. As a universal property, chirality exists widely in nature, and applying it to the explanatory research of deep learning may be helpful to some extent. Inspired by a recent study that used CNN (convolutional neural network), which applied visual chirality, to distinguish whether an image is flipped or not. In this paper, we study feature chirality innovatively, which shows how the statistics of deep learning models' feature data are changed by training. We rethink the feature-level chirality property, propose the feature chirality, and give the measure. Our analysis of feature chirality on AlexNet, VGG, and ResNet reveals similar but surprising results, including the prevalence of feature chirality in these models, the initialization methods of the models do not affect feature chirality. Our work shows that feature chirality implies model evaluation, interpretability of the model, and model parameters optimization.

cs.CV↗

arXiv2023

A Sea-Land Clutter Classification Framework for Over-the-Horizon-Radar Based on Weighted Loss Semi-supervised GAN

Deep convolutional neural network has made great achievements in sea-land clutter classification for over-the-horizon-radar (OTHR). The premise is that a large number of labeled training samples must be provided for a sea-land clutter classifier. In practical engineering applications, it is relatively easy to obtain label-free sea-land clutter samples. However, the labeling process is extremely cumbersome and requires expertise in the field of OTHR. To solve this problem, we propose an improved generative adversarial network, namely weighted loss semi-supervised generative adversarial network (WL-SSGAN). Specifically, we propose a joint feature matching loss by weighting the middle layer features of the discriminator of semi-supervised generative adversarial network. Furthermore, we propose the weighted loss of WL-SSGAN by linearly weighting standard adversarial loss and joint feature matching loss. The semi-supervised classification performance of WL-SSGAN is evaluated on a sea-land clutter dataset. The experimental results show that WL-SSGAN can improve the performance of the fully supervised classifier with only a small number of labeled samples by utilizing a large number of unlabeled sea-land clutter samples. Further, the proposed weighted loss is superior to both the adversarial loss and the feature matching loss. Additionally, we compare WL-SSGAN with conventional semi-supervised classification methods and demonstrate that WL-SSGAN achieves the highest classification accuracy.

cs.CV↗

arXiv2023

Intermediate complex structure limit for Calabi-Yau metrics

We study polarised degenerations of $n$-dimensional Calabi-Yau hypersurfaces $\{ F_0F_1\ldots F_m+tF=0\} $, where the essential skeleton has dimension $1\leq m\leq n-1$. We will describe the limiting behaviour of the Calabi-Yau potential at the $C^0$-level, in terms of an optimal transport problem, and in terms of non-archimedean potential theory.

math.DG↗

arXiv2023

MH-DETR: Video Moment and Highlight Detection with Cross-modal Transformer

With the increasing demand for video understanding, video moment and highlight detection (MHD) has emerged as a critical research topic. MHD aims to localize all moments and predict clip-wise saliency scores simultaneously. Despite progress made by existing DETR-based methods, we observe that these methods coarsely fuse features from different modalities, which weakens the temporal intra-modal context and results in insufficient cross-modal interaction. To address this issue, we propose MH-DETR (Moment and Highlight Detection Transformer) tailored for MHD. Specifically, we introduce a simple yet efficient pooling operator within the uni-modal encoder to capture global intra-modal context. Moreover, to obtain temporally aligned cross-modal features, we design a plug-and-play cross-modal interaction module between the encoder and decoder, seamlessly integrating visual and textual features. Comprehensive experiments on QVHighlights, Charades-STA, Activity-Net, and TVSum datasets show that MH-DETR outperforms existing state-of-the-art methods, demonstrating its effectiveness and superiority. Our code is available at https://github.com/YoucanBaby/MH-DETR.

cs.CV↗

arXiv2023

Defect emission and its dipole orientation in layered ternary Znln2S4 semiconductor

Defect engineering is promising to tailor the physical properties of two-dimensional (2D) semiconductors for function-oriented electronics and optoelectronics. Compared with the extensively studied 2D binary materials, the origin of defects and their influence on physical properties of 2D ternary semiconductors have not been clarified. In this work, we thoroughly studied the effect of defects on the electronic structure and optical properties of few-layer hexagonal Znln2S4 via versatile spectroscopic tools in combination with theoretical calculations. It has been demonstrated that the Zn-In anti-structural defects induce the formation of a series of donor and acceptor levels inside the bandgap, leading to rich recombination paths for defect emission and extrinsic absorption. Impressively, the emission of donor-acceptor pair (DAP) in Znln2S4 can be significantly tailored by electrostatic gating due to efficient tunability of Fermi level (Ef). Furthermore, the layer-dependent dipole orientation of defect emission in Znln2S4 was directly revealed by back focal plane (BFP) imagining, where it presents obviously in-plane dipole orientation within a dozen layers thickness of Znln2S4. These unique features of defects in Znln2S4 including extrinsic absorption, rich recombination paths, gate tunability and in-plane dipole orientation will definitely benefit to the advanced orientation-functional optoelectronic applications.

cond-mat.mtrl-sci↗

arXiv2023

Binary stochasticity enabled highly efficient neuromorphic deep learning achieves better-than-software accuracy

Deep learning needs high-precision handling of forwarding signals, backpropagating errors, and updating weights. This is inherently required by the learning algorithm since the gradient descent learning rule relies on the chain product of partial derivatives. However, it is challenging to implement deep learning in hardware systems that use noisy analog memristors as artificial synapses, as well as not being biologically plausible. Memristor-based implementations generally result in an excessive cost of neuronal circuits and stringent demands for idealized synaptic devices. Here, we demonstrate that the requirement for high precision is not necessary and that more efficient deep learning can be achieved when this requirement is lifted. We propose a binary stochastic learning algorithm that modifies all elementary neural network operations, by introducing (i) stochastic binarization of both the forwarding signals and the activation function derivatives, (ii) signed binarization of the backpropagating errors, and (iii) step-wised weight updates. Through an extensive hybrid approach of software simulation and hardware experiments, we find that binary stochastic deep learning systems can provide better performance than the software-based benchmarks using the high-precision learning algorithm. Also, the binary stochastic algorithm strongly simplifies the neural network operations in hardware, resulting in an improvement of the energy efficiency for the multiply-and-accumulate operations by more than three orders of magnitudes.

cs.NE↗

arXiv2023

Coordinated Control of Path Tracking and Yaw Stability for Distributed Drive Electric Vehicle Based on AMPC and DYC

Maintaining both path-tracking accuracy and yaw stability of distributed drive electric vehicles (DDEVs) under various driving conditions presents a significant challenge in the field of vehicle control. To address this limitation, a coordinated control strategy that integrates adaptive model predictive control (AMPC) path-tracking control and direct yaw moment control (DYC) is proposed for DDEVs. The proposed strategy, inspired by a hierarchical framework, is coordinated by the upper layer of path-tracking control and the lower layer of direct yaw moment control. Based on the linear time-varying model predictive control (LTV MPC) algorithm, the effects of prediction horizon and weight coefficients on the path-tracking accuracy and yaw stability of the vehicle are compared and analyzed first. According to the aforementioned analysis, an AMPC path-tracking controller with variable prediction horizon and weight coefficients is designed considering the vehicle speed's variation in the upper layer. The lower layer involves DYC based on the linear quadratic regulator (LQR) technique. Specifically, the intervention rule of DYC is determined by the threshold of the yaw rate error and the phase diagram of the sideslip angle. Extensive simulation experiments are conducted to evaluate the proposed coordinated control strategy under different driving conditions. The results show that, under variable speed and low adhesion conditions, the vehicle's yaw stability and path-tracking accuracy have been improved by 21.58\% and 14.43\%, respectively, compared to AMPC. Similarly, under high speed and low adhesion conditions, the vehicle's yaw stability and path-tracking accuracy have been improved by 44.30\% and 14.25\%, respectively, compared to the coordination of LTV MPC and DYC. The results indicate that the proposed adaptive path-tracking controller is effective across different speeds.

cs.RO↗

arXiv2023

An Unsupervised STDP-based Spiking Neural Network Inspired By Biologically Plausible Learning Rules and Connections

The backpropagation algorithm has promoted the rapid development of deep learning, but it relies on a large amount of labeled data and still has a large gap with how humans learn. The human brain can quickly learn various conceptual knowledge in a self-organized and unsupervised manner, accomplished through coordinating various learning rules and structures in the human brain. Spike-timing-dependent plasticity (STDP) is a general learning rule in the brain, but spiking neural networks (SNNs) trained with STDP alone is inefficient and perform poorly. In this paper, taking inspiration from short-term synaptic plasticity, we design an adaptive synaptic filter and introduce the adaptive spiking threshold as the neuron plasticity to enrich the representation ability of SNNs. We also introduce an adaptive lateral inhibitory connection to adjust the spikes balance dynamically to help the network learn richer features. To speed up and stabilize the training of unsupervised spiking neural networks, we design a samples temporal batch STDP (STB-STDP), which updates weights based on multiple samples and moments. By integrating the above three adaptive mechanisms and STB-STDP, our model greatly accelerates the training of unsupervised spiking neural networks and improves the performance of unsupervised SNNs on complex tasks. Our model achieves the current state-of-the-art performance of unsupervised STDP-based SNNs in the MNIST and FashionMNIST datasets. Further, we tested on the more complex CIFAR10 dataset, and the results fully illustrate the superiority of our algorithm. Our model is also the first work to apply unsupervised STDP-based SNNs to CIFAR10. At the same time, in the small-sample learning scenario, it will far exceed the supervised ANN using the same structure.

cs.NE↗

arXiv2023

GammaE: Gamma Embeddings for Logical Queries on Knowledge Graphs

Embedding knowledge graphs (KGs) for multi-hop logical reasoning is a challenging problem due to massive and complicated structures in many KGs. Recently, many promising works projected entities and queries into a geometric space to efficiently find answers. However, it remains challenging to model the negation and union operator. The negation operator has no strict boundaries, which generates overlapped embeddings and leads to obtaining ambiguous answers. An additional limitation is that the union operator is non-closure, which undermines the model to handle a series of union operators. To address these problems, we propose a novel probabilistic embedding model, namely Gamma Embeddings (GammaE), for encoding entities and queries to answer different types of FOL queries on KGs. We utilize the linear property and strong boundary support of the Gamma distribution to capture more features of entities and queries, which dramatically reduces model uncertainty. Furthermore, GammaE implements the Gamma mixture method to design the closed union operator. The performance of GammaE is validated on three large logical query datasets. Experimental results show that GammaE significantly outperforms state-of-the-art models on public benchmarks.

cs.LG↗

arXiv2023

SSBM: A Signed Stochastic Block Model for Multiple Structure Discovery in Large-Scale Exploratory Signed Networks

Signed network structure discovery has received extensive attention and has become a research focus in the field of network science. However, most of the existing studies are focused on the networks with a single structure, e.g., community or bipartite, while ignoring multiple structures, e.g., the coexistence of community and bipartite structures. Furthermore, existing studies were faced with challenge regarding large-scale signed networks due to their high time complexity, especially when determining the number of clusters in the observed network without any prior knowledge. In view of this, we propose a mathematically principled method for signed network multiple structure discovery named the Signed Stochastic Block Model (SSBM). The SSBM can capture the multiple structures contained in signed networks, e.g., community, bipartite, and coexistence of them, by adopting a probabilistic model. Moreover, by integrating the minimum message length (MML) criterion and component-wise EM (CEM) algorithm, a scalable learning algorithm that has the ability of model selection is proposed to handle large-scale signed networks. By comparing state-of-the-art methods on synthetic and real-world signed networks, extensive experimental results demonstrate the effectiveness and efficiency of SSBM in discovering large-scale exploratory signed networks with multiple structures.

cs.SI↗

arXiv2023

Solving the domain wall problem with first-order phase transition

Domain wall networks are two-dimensional topological defects generally predicted in many beyond standard model physics. In this Letter, we propose to solve the domain wall problem with the first-order phase transition. We numerically study the phase transition dynamics, and for the first time show that the domain walls reached scaling regime can be diluted through the interaction with vacuum bubbles during the first-order phase transition. We find that the amplitude of the gravitational waves produced by the second-stage first-order phase transition is several orders higher than that from the domain walls evolution in the scaling regime. The scale of the first-order phase transition that dilute the domain walls can be probed through gravitational waves detection.

hep-ph↗

arXiv2023

Pareto-optimal lane-changing motion planning in mixed traffic

This paper applies the pareto-optimal concept to LC (lane-changing) motion planning in the presence of mixed traffic including manual and autonomous vehicles. Firstly, a multiobjective optimization problem is presented, in which the comfort, efficiency and safety of the LC vehicle and the surrounding vehicles are jointly modelled. Thereafter, the pareto-optimal solutions are obtained through employing the NSGA-II (Non-dominated Sorting Genetic -II) algorithm. Finally, the experiment section analyzes the (macroscopic and microscopic) lane-changing impact from a pareto-optimal perspective. Also, a comprehensive sensitivity analysis is conducted. Our results demonstrate that our algorithm could significantly reduce the lane-changing impact within its region, and the total costs are reduced in the range of 10.94% to 48.66%. This paper could be considered as a preliminary research framework for the application of the pareto-optimal concept. We hope this research will provide valuable insights into autonomous driving technology.

cs.RO↗

arXiv2023

Multi-Microgrid Collaborative Optimization Scheduling Using an Improved Multi-Agent Soft Actor-Critic Algorithm

The implementation of a multi-microgrid (MMG) system with multiple renewable energy sources enables the facilitation of electricity trading. To tackle the energy management problem of a MMG system, which consists of multiple renewable energy microgrids belonging to different operating entities, this paper proposes a MMG collaborative optimization scheduling model based on a multi-agent centralized training distributed execution framework. To enhance the generalization ability of dealing with various uncertainties, we also propose an improved multi-agent soft actor-critic (MASAC) algorithm, which facilitates en-ergy transactions between multi-agents in MMG, and employs automated machine learning (AutoML) to optimize the MASAC hyperparameters to further improve the generalization of deep reinforcement learning (DRL). The test results demonstrate that the proposed method successfully achieves power complementarity between different entities, and reduces the MMG system operating cost. Additionally, the proposal significantly outperforms other state-of-the-art reinforcement learning algorithms with better economy and higher calculation efficiency.

eess.SY↗

arXiv2023

Evaluating the principle of relatedness: Estimation, drivers and implications for policy

A growing body of research documents that the size and growth of an industry in a place depends on how much related activity is found there. This fact is commonly referred to as the "principle of relatedness". However, there is no consensus on why we observe the principle of relatedness, how best to determine which industries are related or how this empirical regularity can help inform local industrial policy. We perform a structured search over tens of thousands of specifications to identify robust -- in terms of out-of-sample predictions -- ways to determine how well industries fit the local economies of US cities. To do so, we use data that allow us to derive relatedness from observing which industries co-occur in the portfolios of establishments, firms, cities and countries. Different portfolios yield different relatedness matrices, each of which help predict the size and growth of local industries. However, our specification search not only identifies ways to improve the performance of such predictions, but also reveals new facts about the principle of relatedness and important trade-offs between predictive performance and interpretability of relatedness patterns. We use these insights to deepen our theoretical understanding of what underlies path-dependent development in cities and expand existing policy frameworks that rely on inter-industry relatedness analysis.

physics.soc-ph↗