arXiv subjects

Li Ning

Publications and source records attributed to Li Ning.

12 recordsLinked to original sources

arXiv2026

QeHDC: Hyperdimensional Computing based on Quantum-enhanced binding and SuperClass Construction

Hyperdimensional Computing (HDC) is a robust computational framework inspired by human cognition characterized by simple and efficient operations within high-dimensional vector spaces. Quantum-enhanced Hyperdimensional Computing (QeHDC) extends classical HDC by leveraging quantum mechanical properties to enhance computational efficiency. In this paper, we propose a novel Quantum HDC framework featuring a one-pass training method, leveraging sinusoidal and quantum encoding to project classical data into quantum amplitude states efficiently. Our framework introduces an innovative reference-state-based quantum binding operation realized via quantum circuits. Furthermore, we propose a density-matrix-based superclass generation strategy employing eigenvalue decomposition to extract critical quantum state features effectively, enabling a more accurate and robust class representation. Experimental evaluations conducted on standard benchmark datasets demonstrate our approach's superior performance, robustness to noise, and computational feasibility compared to traditional classical and existing quantum-enhanced approaches. The results highlight the practical benefits and potential of Quantum HDC for quantum-enhanced classification tasks and pave the way for future advancements in quantum-inspired computational paradigms.

cs.LG↗

arXiv2025

ELITE: Embedding-Less retrieval with Iterative Text Exploration

Large Language Models (LLMs) have achieved impressive progress in natural language processing, but their limited ability to retain long-term context constrains performance on document-level or multi-turn tasks. Retrieval-Augmented Generation (RAG) mitigates this by retrieving relevant information from an external corpus. However, existing RAG systems often rely on embedding-based retrieval trained on corpus-level semantic similarity, which can lead to retrieving content that is semantically similar in form but misaligned with the question's true intent. Furthermore, recent RAG variants construct graph- or hierarchy-based structures to improve retrieval accuracy, resulting in significant computation and storage overhead. In this paper, we propose an embedding-free retrieval framework. Our method leverages the logical inferencing ability of LLMs in retrieval using iterative search space refinement guided by our novel importance measure and extend our retrieval results with logically related information without explicit graph construction. Experiments on long-context QA benchmarks, including NovelQA and Marathon, show that our approach outperforms strong baselines while reducing storage and runtime by over an order of magnitude.

cs.CL↗

arXiv2023

Molecular design method based on novel molecular representation and variational auto-encoder

Based on the traditional VAE, a novel neural network model is presented, with the latest molecular representation, SELFIES, to improve the effect of generating new molecules. In this model, multi-layer convolutional network and Fisher information are added to the original encoding layer to learn the data characteristics and guide the encoding process, which makes the features of the data hiding layer more aggregated, and integrates the Long Short Term Memory neural network (LSTM) into the decoding layer for better data generation, which effectively solves the degradation phenomenon generated by the encoding layer and decoding layer of the original VAE model. Through experiments on zinc molecular data sets, it is found that the similarity in the new VAE is 8.47% higher than that of the original ones. SELFIES are better at generating a variety of molecules than the traditional molecular representation, SELFIES. Experiments have shown that using SELFIES and the new VAE model presented in this paper can improve the effectiveness of generating new molecules.

q-bio.BM↗

arXiv2019

LAC-Nav: Collision-Free Mutiagent Navigation Based on The Local Action Cells

Collision avoidance is one of the most primary requirement in the decentralized multiagent navigations: while the agents are moving towards their own targets, attentions should be paid to avoid the collisions with the others. In this paper, we introduce the concept of local action cell, which provides for each agent a set of velocities that are safe to perform. Based on the realtime updated local action cells, we propose the LAC-Nav approach to navigate the agent with the properly selected velocity; and furthermore, we coupled the local action cell with an adaptive learning framework, in which the effect of selections are evaluated and used as the references for making decisions in the following updates. Through the experiments for three commonly considered scenarios, we demonstrated the efficiency of the proposed approaches, with the comparison to several widely studied strategies.

cs.MA↗

arXiv2019

PyramNet: Point Cloud Pyramid Attention Network and Graph Embedding Module for Classification and Segmentation

With the tide of artificial intelligence, we try to apply deep learning to understand 3D data. Point cloud is an important 3D data structure, which can accurately and directly reflect the real world. In this paper, we propose a simple and effective network, which is named PyramNet, suites for point cloud object classification and semantic segmentation in 3D scene. We design two new operators: Graph Embedding Module(GEM) and Pyramid Attention Network(PAN). Specifically, GEM projects point cloud onto the graph and practices the covariance matrix to explore the relationship between points, so as to improve the local feature expression ability of the model. PAN assigns some strong semantic features to each point to retain fine geometric features as much as possible. Furthermore, we provide extensive evaluation and analysis for the effectiveness of PyramNet. Empirically, we evaluate our model on ModelNet40, ShapeNet and S3DIS.

cs.CV↗

arXiv2019

Cognitive computation of brain disorders based primarily on ocular responses

The present review presents multiple techniques in which ocular assessments may serve as a noninvasive approach for the early diagnoses of various cognitive and psychiatric disorders, such as Alzheimer's disease (AD), autism spectrum disorder (ASD), schizophrenia (SZ), and major depressive disorder (MDD). Real-time ocular responses are tightly associated with emotional and cognitive processing within the central nervous system. Patterns seen in saccades, pupillary responses, and blinking, as well as retinal microvasculature and morphology visualized via office-based ophthalmic imaging, are potential biomarkers for the screening and evaluation of cognitive and psychiatric disorders. Additionally, rapid advances in artificial intelligence (AI) present a growing opportunity to use machine-learning-based AI, especially deep-learning neural networks, to shed new light on the field of cognitive neuroscience, which may lead to novel evaluations and interventions via ocular approaches for cognitive and psychiatric disorders.

q-bio.NC↗

arXiv2015

Influence Maximization under The Non-progressive Linear Threshold Model

In the problem of influence maximization in information networks, the objective is to choose a set of initially active nodes subject to some budget constraints such that the expected number of active nodes over time is maximized. The linear threshold model has been introduced to study the opinion cascading behavior, for instance, the spread of products and innovations. In this paper, we we extends the classic linear threshold model [18] to capture the non-progressive be- havior. The information maximization problem under our model is proved to be NP-Hard, even for the case when the underlying network has no directed cycles. The first result of this paper is negative. In general, the objective function of the extended linear threshold model is no longer submodular, and hence the hill climbing approach that is commonly used in the existing studies is not applicable. Next, as the main result of this paper, we prove that if the underlying information network is directed acyclic, the objective function is submodular (and monotone). Therefore, in directed acyclic networks with a specified budget we can achieve 1/2 -approximation on maximizing the number of active nodes over a certain period of time by a deterministic algorithm, and achieve the (1 - 1/e )-approximation by a randomized algorithm.

cs.SI↗

arXiv2015

Uniform Information Exchange in Multi-channel Wireless Ad Hoc Networks

In the information exchange problem, k packets that are initially maintained by k nodes need to be disseminated to the whole network as quickly as possible. We consider this problem in single-hop multi- channel networks of n nodes, and propose a uniform protocol that with high probability accomplishes the dissemination in O(k/F + F \cdot log n) rounds, assuming F available channels and collision detection. This result is asymptotically optimal when k is large (k \geq F^2 \cdot log n). To our knowledge, this is the first uniform protocol for information exchange in multi-channel networks.

cs.DC↗

arXiv2012

A 4-state solution to the Firing Squad Synchronization Problem based on hybrid rule 60 and 102 cellular automata

In this paper, we present a 4-state solution to the Firing Squad Synchronization Problem (FSSP) based on hybrid rule 60/102 Cellular Automata(CA). This solution solves the problem on the line of length 2^n with two generals. Previous work on FSSP for 4-state systems focused mostly on linear cellular automata, where synchronizes an infinite number of lines but not all possible lines. We give time-optimal solutions to synchronize an infinite number of lines by rule 60 and rule 102 respectively, and construct a hybrid rule 60 and 102 states transition table. Compared to the known solutions of cellular automata, the hybrid CA way is simpler and faster, the minimal time is (n-1) step.

nlin.CG↗

arXiv2012

Incubators vs Zombies: Fault-Tolerant, Short, Thin and Lanky Spanners for Doubling Metrics

Recently Elkin and Solomon gave a construction of spanners for doubling metrics that has constant maximum degree, hop-diameter O(log n) and lightness O(log n) (i.e., weight O(log n)w(MST). This resolves a long standing conjecture proposed by Arya et al. in a seminal STOC 1995 paper. However, Elkin and Solomon's spanner construction is extremely complicated; we offer a simple alternative construction that is very intuitive and is based on the standard technique of net tree with cross edges. Indeed, our approach can be readily applied to our previous construction of k-fault tolerant spanners (ICALP 2012) to achieve k-fault tolerance, maximum degree O(k^2), hop-diameter O(log n) and lightness O(k^3 log n).

cs.DS↗

arXiv2012

(1+epsilon)-Distance Oracle for Planar Labeled Graph

Given a vertex-labeled graph, each vertex $v$ is attached with a label from a set of labels. The vertex-label query desires the length of the shortest path from the given vertex to the set of vertices with the given label. We show how to construct an oracle if the given graph is planar, such that $O(\frac{1}εn\log n)$ storing space is needed, and any vertex-label query could be answered in $O(\frac{1}ε\log n\log ρ)$ time with stretch $1+ε$. $ρ$ is the radius of the given graph, which is half of the diameter. For the case that $ρ= O(\log n)$, we construct an oracle that achieves $O(\log n)$ query time, without changing the order of storing space.

cs.DS↗

arXiv1993

Rotating Rayleigh-Bénard Convection: Aspect Ratio Dependence of the Initial Bifurcations

The initial bifurcations in rotating Rayleigh-Bénard convection are studied in the range of dimensionless rotation rate $0 < Ω< 2150$ for an aspect-ratio-2.5 cylindrical cell. We used simultaneous optical shadowgraph, heat transport and local temperature measurements to determine the stability and characteristics of the azimuthally-periodic wall convection state. We also show that the second transition corresponds to the onset of bulk convection. Our results for critical Rayleigh numbers, precession frequencies and critical mode numbers agree well with theoretical results. The dynamics of the wall convection state can be described by a complex Ginzburg-Landau amplitude equation.

patt-sol↗