Searcharxiv⌕ Search

arXiv subjects

Wei Fang

Publications and source records attributed to Wei Fang.

At least 55 records · Page 3Linked to original sources
arXiv2024

Parallel Spiking Neurons with High Efficiency and Ability to Learn Long-term Dependencies

Vanilla spiking neurons in Spiking Neural Networks (SNNs) use charge-fire-reset neuronal dynamics, which can only be simulated serially and can hardly learn long-time dependencies. We find that when removing reset, the neuronal dynamics can be reformulated in a non-iterative form and parallelized. By rewriting neuronal dynamics without reset to a general formulation, we propose the Parallel Spiking Neuron (PSN), which generates hidden states that are independent of their predecessors, resulting in parallelizable neuronal dynamics and extremely high simulation speed. The weights of inputs in the PSN are fully connected, which maximizes the utilization of temporal information. To avoid the use of future inputs for step-by-step inference, the weights of the PSN can be masked, resulting in the masked PSN. By sharing weights across time-steps based on the masked PSN, the sliding PSN is proposed to handle sequences of varying lengths. We evaluate the PSN family on simulation speed and temporal/static data classification, and the results show the overwhelming advantage of the PSN family in efficiency and accuracy. To the best of our knowledge, this is the first study about parallelizing spiking neurons and can be a cornerstone for the spiking deep learning research. Our codes are available at \url{https://github.com/fangwei123456/Parallel-Spiking-Neuron}.

cs.NE↗
arXiv2023

SpikingJelly: An open-source machine learning infrastructure platform for spike-based intelligence

Spiking neural networks (SNNs) aim to realize brain-inspired intelligence on neuromorphic chips with high energy efficiency by introducing neural dynamics and spike properties. As the emerging spiking deep learning paradigm attracts increasing interest, traditional programming frameworks cannot meet the demands of the automatic differentiation, parallel computation acceleration, and high integration of processing neuromorphic datasets and deployment. In this work, we present the SpikingJelly framework to address the aforementioned dilemma. We contribute a full-stack toolkit for pre-processing neuromorphic datasets, building deep SNNs, optimizing their parameters, and deploying SNNs on neuromorphic chips. Compared to existing methods, the training of deep SNNs can be accelerated $11\times$, and the superior extensibility and flexibility of SpikingJelly enable users to accelerate custom models at low costs through multilevel inheritance and semiautomatic code generation. SpikingJelly paves the way for synthesizing truly energy-efficient SNN-based machine intelligence systems, which will enrich the ecology of neuromorphic computing.

cs.NE↗
arXiv2023

CancerUniT: Towards a Single Unified Model for Effective Detection, Segmentation, and Diagnosis of Eight Major Cancers Using a Large Collection of CT Scans

Human readers or radiologists routinely perform full-body multi-organ multi-disease detection and diagnosis in clinical practice, while most medical AI systems are built to focus on single organs with a narrow list of a few diseases. This might severely limit AI's clinical adoption. A certain number of AI models need to be assembled non-trivially to match the diagnostic process of a human reading a CT scan. In this paper, we construct a Unified Tumor Transformer (CancerUniT) model to jointly detect tumor existence & location and diagnose tumor characteristics for eight major cancers in CT scans. CancerUniT is a query-based Mask Transformer model with the output of multi-tumor prediction. We decouple the object queries into organ queries, tumor detection queries and tumor diagnosis queries, and further establish hierarchical relationships among the three groups. This clinically-inspired architecture effectively assists inter- and intra-organ representation learning of tumors and facilitates the resolution of these complex, anatomically related multi-organ cancer image reading tasks. CancerUniT is trained end-to-end using a curated large-scale CT images of 10,042 patients including eight major types of cancers and occurring non-cancer tumors (all are pathology-confirmed with 3D tumor masks annotated by radiologists). On the test set of 631 patients, CancerUniT has demonstrated strong performance under a set of clinically relevant evaluation metrics, substantially outperforming both multi-disease methods and an assembly of eight single-organ expert models in tumor detection, segmentation, and diagnosis. This moves one step closer towards a universal high performance cancer screening tool.

eess.IV↗
arXiv2023

Analyzing and controlling diversity in quantum-behaved particle swarm optimization

This paper addresses the issues of controlling and analyzing the population diversity in quantum-behaved particle swarm optimization (QPSO), which is an optimization approach motivated by concepts in quantum mechanics and PSO. In order to gain an in-depth understanding of the role the diversity plays in the evolving process, we first define the genotype diversity by the distance to the average point of the particles' positions and the phenotype diversity by the fitness values for the QPSO. Then, the correlations between the two types of diversities and the search performance are tested and analyzed on several benchmark functions, and the distance-to-average-point diversity is showed to have stronger association with the search performance during the evolving processes. Finally, in the light of the performed diversity analyses, two strategies for controlling the distance-to-average-point diversities are proposed for the purpose of improving the search ability of the QPSO algorithm. Empirical studies on the QPSO with the introduced diversity control methods are performed on a set of benchmark functions from the CEC 2005 benchmark suite. The performance of the proposed methods are evaluated and compared with the original QPSO and other PSO variants.

cs.NE↗
arXiv2023

Coherent control of a high-orbital hole in a semiconductor quantum dot

Coherently driven semiconductor quantum dots are one of the most promising platforms for non-classical light sources and quantum logic gates which form the foundation of photonic quantum technologies. However, to date, coherent manipulation of single charge carriers in quantum dots is limited mainly to their lowest orbital states. Ultrafast coherent control of high-orbital states is obstructed by the demand for tunable terahertz pulses. To break this constraint, we demonstrate an all-optical method to control high-orbital states of a hole via stimulated Auger process. The coherent nature of the Auger process is proved by Rabi oscillation and Ramsey interference. Harnessing this coherence further enables the investigation of single-hole relaxation mechanism. A hole relaxation time of 161 ps is observed and attributed to the phonon bottleneck effect. Our work opens new possibilities for understanding the fundamental properties of high-orbital states in quantum emitters and developing new types of orbital-based quantum photonic devices.

quant-ph↗
arXiv2023

Auto-Spikformer: Spikformer Architecture Search

The integration of self-attention mechanisms into Spiking Neural Networks (SNNs) has garnered considerable interest in the realm of advanced deep learning, primarily due to their biological properties. Recent advancements in SNN architecture, such as Spikformer, have demonstrated promising outcomes by leveraging Spiking Self-Attention (SSA) and Spiking Patch Splitting (SPS) modules. However, we observe that Spikformer may exhibit excessive energy consumption, potentially attributable to redundant channels and blocks. To mitigate this issue, we propose Auto-Spikformer, a one-shot Transformer Architecture Search (TAS) method, which automates the quest for an optimized Spikformer architecture. To facilitate the search process, we propose methods Evolutionary SNN neurons (ESNN), which optimizes the SNN parameters, and apply the previous method of weight entanglement supernet training, which optimizes the Vision Transformer (ViT) parameters. Moreover, we propose an accuracy and energy balanced fitness function $\mathcal{F}_{AEB}$ that jointly considers both energy consumption and accuracy, and aims to find a Pareto optimal combination that balances these two objectives. Our experimental results demonstrate the effectiveness of Auto-Spikformer, which outperforms the state-of-the-art method including CNN or ViT models that are manually or automatically designed while significantly reducing energy consumption.

cs.NE↗
arXiv2023

Expand, Rerank, and Retrieve: Query Reranking for Open-Domain Question Answering

We propose EAR, a query Expansion And Reranking approach for improving passage retrieval, with the application to open-domain question answering. EAR first applies a query expansion model to generate a diverse set of queries, and then uses a query reranker to select the ones that could lead to better retrieval results. Motivated by the observation that the best query expansion often is not picked by greedy decoding, EAR trains its reranker to predict the rank orders of the gold passages when issuing the expanded queries to a given retriever. By connecting better the query expansion model and retriever, EAR significantly enhances a traditional sparse retrieval method, BM25. Empirically, EAR improves top-5/20 accuracy by 3-8 and 5-10 points in in-domain and out-of-domain settings, respectively, when compared to a vanilla query expansion model, GAR, and a dense retrieval model, DPR.

cs.CL↗
arXiv2023

Temporal Contrastive Learning for Spiking Neural Networks

Biologically inspired spiking neural networks (SNNs) have garnered considerable attention due to their low-energy consumption and spatio-temporal information processing capabilities. Most existing SNNs training methods first integrate output information across time steps, then adopt the cross-entropy (CE) loss to supervise the prediction of the average representations. However, in this work, we find the method above is not ideal for the SNNs training as it omits the temporal dynamics of SNNs and degrades the performance quickly with the decrease of inference time steps. One tempting method to model temporal correlations is to apply the same label supervision at each time step and treat them identically. Although it can acquire relatively consistent performance across various time steps, it still faces challenges in obtaining SNNs with high performance. Inspired by these observations, we propose Temporal-domain supervised Contrastive Learning (TCL) framework, a novel method to obtain SNNs with low latency and high performance by incorporating contrastive supervision with temporal domain information. Contrastive learning (CL) prompts the network to discern both consistency and variability in the representation space, enabling it to better learn discriminative and generalizable features. We extend this concept to the temporal domain of SNNs, allowing us to flexibly and fully leverage the correlation between representations at different time steps. Furthermore, we propose a Siamese Temporal-domain supervised Contrastive Learning (STCL) framework to enhance the SNNs via augmentation, temporal and class constraints simultaneously. Extensive experimental results demonstrate that SNNs trained by our TCL and STCL can achieve both high performance and low latency, achieving state-of-the-art performance on a variety of datasets (e.g., CIFAR-10, CIFAR-100, and DVS-CIFAR10).

cs.CV↗
arXiv2023

Composite picosecond control of atomic state through a nanofiber interface

Atoms are ideal quantum sensors and quantum light emitters. Interfacing atoms with nanophotonic devices promises novel nanoscale sensing and quantum optical functionalities. But precise optical control of atomic states in these devices is challenged by the spatially varying light-atom coupling strength, generic to nanophotonic. We demonstrate numerically that despite the inhomogenuity, composite picosecond optical pulses with optimally tailored phases are able to evanescently control the atomic electric dipole transitions nearly perfectly, with $f>99\%$ fidelity across large enough volumes for {\it e.g.} controlling cold atoms confined in near-field optical lattices. Our proposal is followed by a proof-of-principle demonstration with a $^{85}$Rb vapor -- optical nanofiber interface, where the excitation by an $N=3$ sequence of guided picosecond D1 control reduces the absorption of a co-guided nanosecond D2 probe by up to $\sim70\%$. The close-to-ideal performance is corroborated by comparing the absorption data across the parameter space with first-principle modeling of the mesoscopic atomic vapor response. Extension of the composite technique to $N\geq 5$ appears highly feasible to support arbitrary local control of atomic dipoles with exquisite precision. This unprecedented ability would allow error-resilient atomic spectroscopy and open up novel nonlinear quantum optical research with atom-nanophotonic interfaces.

physics.optics↗
arXiv2023

Robust Gaussian Process Regression method for efficient reaction pathway optimization: application to surface processes

Simulation of surface processes is a key part of computational chemistry that offers atomic-scale insights into mechanisms of heterogeneous catalysis, diffusion dynamics, as well as quantum tunneling phenomena. The most common theoretical approaches involve optimization of reaction pathways, including semiclassical tunneling pathways (called instantons). However, the computational effort can be demanding, especially for instanton optimizations with ab initio electronic structure. Recently, machine learning has been applied to accelerate reaction-pathway optimization, showing great potential for a wide range of applications. However, previous methods suffer from practical issues such as unfavorable scaling with respect to the size of the descriptor, and were mostly designed for reactions in the gas phase. We propose an improved framework based on Gaussian process regression for general transformed coordinates, which can alleviate the size problem. The descriptor combines internal and Cartesian coordinates, which improves the performance for modeling surface processes. We demonstrate with eleven instanton optimizations in three example systems that the new approach makes ab initio instanton optimization significantly cheaper, such that it becomes not much more expensive than a classical transition-state theory calculation.

physics.chem-ph↗
arXiv2023

Interpretable Unified Language Checking

Despite recent concerns about undesirable behaviors generated by large language models (LLMs), including non-factual, biased, and hateful language, we find LLMs are inherent multi-task language checkers based on their latent representations of natural and social knowledge. We present an interpretable, unified, language checking (UniLC) method for both human and machine-generated language that aims to check if language input is factual and fair. While fairness and fact-checking tasks have been handled separately with dedicated models, we find that LLMs can achieve high performance on a combination of fact-checking, stereotype detection, and hate speech detection tasks with a simple, few-shot, unified set of prompts. With the ``1/2-shot'' multi-task language checking method proposed in this work, the GPT3.5-turbo model outperforms fully supervised baselines on several language tasks. The simple approach and results suggest that based on strong latent knowledge representations, an LLM can be an adaptive and explainable tool for detecting misinformation, stereotypes, and hate speech.

cs.CL↗
arXiv2023

Optimal ANN-SNN Conversion for High-accuracy and Ultra-low-latency Spiking Neural Networks

Spiking Neural Networks (SNNs) have gained great attraction due to their distinctive properties of low power consumption and fast inference on neuromorphic hardware. As the most effective method to get deep SNNs, ANN-SNN conversion has achieved comparable performance as ANNs on large-scale datasets. Despite this, it requires long time-steps to match the firing rates of SNNs to the activation of ANNs. As a result, the converted SNN suffers severe performance degradation problems with short time-steps, which hamper the practical application of SNNs. In this paper, we theoretically analyze ANN-SNN conversion error and derive the estimated activation function of SNNs. Then we propose the quantization clip-floor-shift activation function to replace the ReLU activation function in source ANNs, which can better approximate the activation function of SNNs. We prove that the expected conversion error between SNNs and ANNs is zero, enabling us to achieve high-accuracy and ultra-low-latency SNNs. We evaluate our method on CIFAR-10/100 and ImageNet datasets, and show that it outperforms the state-of-the-art ANN-SNN and directly trained SNNs in both accuracy and time-steps. To the best of our knowledge, this is the first time to explore high-performance ANN-SNN conversion with ultra-low latency (4 time-steps). Code is available at https://github.com/putshua/SNN\_conversion\_QCFS

cs.NE↗
arXiv2023

A Unified Framework for Soft Threshold Pruning

Soft threshold pruning is among the cutting-edge pruning methods with state-of-the-art performance. However, previous methods either perform aimless searching on the threshold scheduler or simply set the threshold trainable, lacking theoretical explanation from a unified perspective. In this work, we reformulate soft threshold pruning as an implicit optimization problem solved using the Iterative Shrinkage-Thresholding Algorithm (ISTA), a classic method from the fields of sparse recovery and compressed sensing. Under this theoretical framework, all threshold tuning strategies proposed in previous studies of soft threshold pruning are concluded as different styles of tuning $L_1$-regularization term. We further derive an optimal threshold scheduler through an in-depth study of threshold scheduling based on our framework. This scheduler keeps $L_1$-regularization coefficient stable, implying a time-invariant objective function from the perspective of optimization. In principle, the derived pruning algorithm could sparsify any mathematical model trained via SGD. We conduct extensive experiments and verify its state-of-the-art performance on both Artificial Neural Networks (ResNet-50 and MobileNet-V1) and Spiking Neural Networks (SEW ResNet-18) on ImageNet datasets. On the basis of this framework, we derive a family of pruning methods, including sparsify-during-training, early pruning, and pruning at initialization. The code is available at https://github.com/Yanqi-Chen/LATS.

cs.LG↗
arXiv2022

Simple Primitives with Feasibility- and Contextuality-Dependence for Open-World Compositional Zero-shot Learning

The task of Compositional Zero-Shot Learning (CZSL) is to recognize images of novel state-object compositions that are absent during the training stage. Previous methods of learning compositional embedding have shown effectiveness in closed-world CZSL. However, in Open-World CZSL (OW-CZSL), their performance tends to degrade significantly due to the large cardinality of possible compositions. Some recent works separately predict simple primitives (i.e., states and objects) to reduce cardinality. However, they consider simple primitives as independent probability distributions, ignoring the heavy dependence between states, objects, and compositions. In this paper, we model the dependence of compositions via feasibility and contextuality. Feasibility-dependence refers to the unequal feasibility relations between simple primitives, e.g., \textit{hairy} is more feasible with \textit{cat} than with \textit{building} in the real world. Contextuality-dependence represents the contextual variance in images, e.g., \textit{cat} shows diverse appearances under the state of \textit{dry} and \textit{wet}. We design Semantic Attention (SA) and generative Knowledge Disentanglement (KD) to learn the dependence of feasibility and contextuality, respectively. SA captures semantics in compositions to alleviate impossible predictions, driven by the visual similarity between simple primitives. KD disentangles images into unbiased feature representations, easing contextual bias in predictions. Moreover, we complement the current compositional probability model with feasibility and contextuality in a compatible format. Finally, we conduct comprehensive experiments to analyze and validate the superior or competitive performance of our model, Semantic Attention and knowledge Disentanglement guided Simple Primitives (SAD-SP), on three widely-used benchmark OW-CZSL datasets.

cs.CV↗
arXiv2022

Neuron with Steady Response Leads to Better Generalization

Regularization can mitigate the generalization gap between training and inference by introducing inductive bias. Existing works have already proposed various inductive biases from diverse perspectives. However, none of them explores inductive bias from the perspective of class-dependent response distribution of individual neurons. In this paper, we conduct a substantial analysis of the characteristics of such distribution. Based on the analysis results, we articulate the Neuron Steadiness Hypothesis: the neuron with similar responses to instances of the same class leads to better generalization. Accordingly, we propose a new regularization method called Neuron Steadiness Regularization (NSR) to reduce neuron intra-class response variance. Based on the Complexity Measure, we theoretically guarantee the effectiveness of NSR for improving generalization. We conduct extensive experiments on Multilayer Perceptron, Convolutional Neural Networks, and Graph Neural Networks with popular benchmark datasets of diverse domains, which show that our Neuron Steadiness Regularization consistently outperforms the vanilla version of models with significant gain and low additional computational overhead.

cs.LG↗
arXiv2022

Modes trimming and clustering in a weakly perturbed high-Q whispering gallery microresonator

In general, a high-Q microresonator can accommodate abundant whispering gallery modes (WGMs) with the mode number increasing with the dimensional sizes of the microresonator. Removing the unnecessary modes while reorganizing the remaining modes is of vital importance, which, however, has been proved challenging and usually results in a tradeoff with the Q of the microresonator. Here, we reveal an effective and controllable mode trimming and clustering mechanism underlying the generation of polygon and star modes in weakly perturbed tapered fiber-coupled lithium niobate whispering gallery microresonators. Experimentally, various polygon and star modes are observed in sequence within a single microresonator by tuning the excitation wavelength or varying the coupling position between a tapered fiber and the circular microresonator, which can be well reproduced with our theoretical model. The finding offers a ubiquitous solution for a broad range of applications requiring elaborate selection and organization of the high-Q WGMs.

physics.optics↗
arXiv2022

General analytical nuclear force and molecular potential energy surface from full configuration interaction quantum Monte Carlo

Full configuration interaction quantum Monte Carlo (FCIQMC) is a state-of-the-art stochastic electronic structure method, providing a methodology to compute FCI-level state energies of molecular systems within a quantum chemical basis. However, especially to probe {\em dynamics} at the FCIQMC level, it is necessary to devise more efficient schemes to produce nuclear forces and potential energy surfaces (PES) from FCIQMC. In this work, we derive the general formula for nuclear force from FCIQMC, and clarify different contributions of the total force. This method to obtain FCIQMC forces eliminates previous restrictions, and can be used with frozen core approximation and free selection of orbitals, making it promising for more efficient nuclear force calculations. After numerical check of this procedure on the binding curve of N$_2$ molecule, we use the FCIQMC energy and force to obtain the full-dimensional ground state PES of water molecule via Gaussian processes regression. The new water FCIQMC PES can be used as the basis for H$_2$O ground state nuclear dynamics, structure optimization, and rotation-vibrational spectrum calculation.

physics.chem-ph↗
arXiv2022

Nonlinear optical radiation of a lithium niobate microcavity

The nonlinear optical radiation of an integrated lithium niobate microcavity is demonstrated, which has been neglected in previous studies of nonlinear photonic devices. We find that the nonlinear coupling between confined optical modes on the chip and continuum modes in free space can be greatly enhanced on the platform of integrated microcavity, with feasible relaxation of the phase-matching condition. With an infrared pump laser, we observe the vertical radiation of second-harmonic wave at the visible band, which indicates a robust phase-matching-free chip-to-free-space frequency converter and also unveils an extra energy dissipation channel for integrated devices. Such an unexpected coherent nonlinear interaction between the free-space beam and the confined mode is also validated by the different frequency generation. Furthermore, based on the phase-matching-free nature of the nonlinear radiation, we build an integrated atomic gas sensor to characterize Rb isotopes with a single telecom laser. The unveiled mechanism of nonlinear optical radiation is universal for all dielectric photonic integrated devices, and provides a simple and robust chip-to-free-space as well as visible-to-telecom interface.

physics.optics↗