arXiv subjects

Tian Chen

Publications and source records attributed to Tian Chen.

At least 37 records · Page 2Linked to original sources

arXiv2024

Progressive Retinal Image Registration via Global and Local Deformable Transformations

Retinal image registration plays an important role in the ophthalmological diagnosis process. Since there exist variances in viewing angles and anatomical structures across different retinal images, keypoint-based approaches become the mainstream methods for retinal image registration thanks to their robustness and low latency. These methods typically assume the retinal surfaces are planar, and adopt feature matching to obtain the homography matrix that represents the global transformation between images. Yet, such a planar hypothesis inevitably introduces registration errors since retinal surface is approximately curved. This limitation is more prominent when registering image pairs with significant differences in viewing angles. To address this problem, we propose a hybrid registration framework called HybridRetina, which progressively registers retinal images with global and local deformable transformations. For that, we use a keypoint detector and a deformation network called GAMorph to estimate the global transformation and local deformable transformation, respectively. Specifically, we integrate multi-level pixel relation knowledge to guide the training of GAMorph. Additionally, we utilize an edge attention module that includes the geometric priors of the images, ensuring the deformation field focuses more on the vascular regions of clinical interest. Experiments on two widely-used datasets, FIRE and FLoRI21, show that our proposed HybridRetina significantly outperforms some state-of-the-art methods. The code is available at https://github.com/lyp-deeplearning/awesome-retinal-registration.

cs.CV↗

arXiv2024

Personalized Quantum Federated Learning for Privacy Image Classification

Quantum federated learning has brought about the improvement of privacy image classification, while the lack of personality of the client model may contribute to the suboptimal of quantum federated learning. A personalized quantum federated learning algorithm for privacy image classification is proposed to enhance the personality of the client model in the case of an imbalanced distribution of images. First, a personalized quantum federated learning model is constructed, in which a personalized layer is set for the client model to maintain the personalized parameters. Second, a personalized quantum federated learning algorithm is introduced to secure the information exchanged between the client and server.Third, the personalized federated learning is applied to image classification on the FashionMNIST dataset, and the experimental results indicate that the personalized quantum federated learning algorithm can obtain global and local models with excellent performance, even in situations where local training samples are imbalanced. The server's accuracy is 100% with 8 clients and a distribution parameter of 100, outperforming the non-personalized model by 7%. The average client accuracy is 2.9% higher than that of the non-personalized model with 2 clients and a distribution parameter of 1. Compared to previous quantum federated learning algorithms, the proposed personalized quantum federated learning algorithm eliminates the need for additional local training while safeguarding both model and data privacy.It may facilitate broader adoption and application of quantum technologies, and pave the way for more secure, scalable, and efficient quantum distribute machine learning solutions.

quant-ph↗

arXiv2024

Frankenstein: Generating Semantic-Compositional 3D Scenes in One Tri-Plane

We present Frankenstein, a diffusion-based framework that can generate semantic-compositional 3D scenes in a single pass. Unlike existing methods that output a single, unified 3D shape, Frankenstein simultaneously generates multiple separated shapes, each corresponding to a semantically meaningful part. The 3D scene information is encoded in one single tri-plane tensor, from which multiple Singed Distance Function (SDF) fields can be decoded to represent the compositional shapes. During training, an auto-encoder compresses tri-planes into a latent space, and then the denoising diffusion process is employed to approximate the distribution of the compositional scenes. Frankenstein demonstrates promising results in generating room interiors as well as human avatars with automatically separated parts. The generated scenes facilitate many downstream applications, such as part-wise re-texturing, object rearrangement in the room or avatar cloth re-targeting. Our project page is available at: https://wolfball.github.io/frankenstein/.

cs.CV↗

arXiv2024

Problem of eigenvalues of stochastic Hamiltonian systems with boundary conditions and Markov chain

In this paper, we study the eigenvalue problem of stochastic Hamiltonian system driven by Brownian motion and Markov chain with boundary conditions and time-dependent coefficients. For any dimensional case, the existence of the first eigenvalue is proven and the corresponding eigenfunctions are constructed by virtue of dual transformation and generalized Riccati equation system. Furthermore, we have more finely characterized the existence of all eigenvalues and constructed the related eigenfunctions for one-dimensional Hamiltonian system. Moreover, the increasing order of these eigenvalues have also been given.

math.PR↗

arXiv2024

Viscosity solutions for mean field optimal switching with a two-time-scale Markov chain

In this paper, we consider the mean field optimal switching problem with a Markov chain under viscosity solution notion. Based on the conditional distribution of the Markov chain, the value function and corresponding dynamic programming principle (DPP) are established. The switching problem is characterized by an obstacle equation on the Wasserstein space, and the existence, stability, and comparison principle are obtained in the sense of viscosity solution. In particular, we consider a two-time-scale structure and obtain the convergence of the limit system. As an application of our theoretical results, an innovative example concerning the stock trading problem in a regime switching market is solved.

math.OC↗

arXiv2024

Dual Structure-Aware Image Filterings for Semi-supervised Medical Image Segmentation

Semi-supervised image segmentation has attracted great attention recently. The key is how to leverage unlabeled images in the training process. Most methods maintain consistent predictions of the unlabeled images under variations (e.g., adding noise/perturbations, or creating alternative versions) in the image and/or model level. In most image-level variation, medical images often have prior structure information, which has not been well explored. In this paper, we propose novel dual structure-aware image filterings (DSAIF) as the image-level variations for semi-supervised medical image segmentation. Motivated by connected filtering that simplifies image via filtering in structure-aware tree-based image representation, we resort to the dual contrast invariant Max-tree and Min-tree representation. Specifically, we propose a novel connected filtering that removes topologically equivalent nodes (i.e. connected components) having no siblings in the Max/Min-tree. This results in two filtered images preserving topologically critical structure. Applying the proposed DSAIF to mutually supervised networks decreases the consensus of their erroneous predictions on unlabeled images. This helps to alleviate the confirmation bias issue of overfitting to noisy pseudo labels of unlabeled images, and thus effectively improves the segmentation performance. Extensive experimental results on three benchmark datasets demonstrate that the proposed method significantly/consistently outperforms some state-of-the-art methods. The source codes will be publicly available.

cs.CV↗

arXiv2023

Flover: A Temporal Fusion Framework for Efficient Autoregressive Model Parallel Inference

Autoregressive models, despite their commendable performance in a myriad of generative tasks, face challenges stemming from their inherently sequential structure. Inference on these models, by design, harnesses a temporal dependency, where the current token's probability distribution is conditioned on preceding tokens. This inherent characteristic severely impedes computational efficiency during inference as a typical inference request can require more than thousands of tokens, where generating each token requires a load of entire model weights, making the inference more memory-bound. The large overhead becomes profound in real deployment where requests arrive randomly, necessitating various generation lengths. Existing solutions, such as dynamic batching and concurrent instances, introduce significant response delays and bandwidth contention, falling short of achieving optimal latency and throughput. To address these shortcomings, we propose Flover -- a temporal fusion framework for efficiently inferring multiple requests in parallel. We deconstruct the general generation pipeline into pre-processing and token generation, and equip the framework with a dedicated work scheduler for fusing the generation process temporally across all requests. By orchestrating the token-level parallelism, Flover exhibits optimal hardware efficiency and significantly spares the system resources. By further employing a fast buffer reordering algorithm that allows memory eviction of finished tasks, it brings over 11x inference speedup on GPT and 16x on LLAMA compared to the cutting-edge solutions provided by NVIDIA FasterTransformer. Crucially, by leveraging the advanced tensor parallel technique, Flover proves efficacious across diverse computational landscapes, from single-GPU setups to distributed scenarios, thereby offering robust performance optimization that adapts to variable use cases.

cs.DC↗

arXiv2023

Topologically protected entanglement switching around exceptional points

The robust operation of quantum entanglement states are crucial for applications in quantum information, computing, and communications1-3. However, it has always been a great challenge to complete such a task because of decoherence and disorder. Here, we propose theoretically and demonstrate experimentally an effective scheme to realize robust operation of quantum entanglement states by designing quadruple degeneracy exceptional points. By encircling the exceptional points on two overlapping Riemann energy surfaces, we have realized a chiral switch for entangled states with high fidelity. Owing to the topological protection conferred by the Riemann surface structure, this switching of chirality exhibits strong robustness against perturbations in the encircling path. Furthermore, we have experimentally validated such a scheme on a quantum walk platform. Our work opens up a new way for the application of non-Hermitian physics in the field of quantum information.

quant-ph↗

arXiv2023

Unified High-binding Watermark for Unconditional Image Generation Models

Deep learning techniques have implemented many unconditional image generation (UIG) models, such as GAN, Diffusion model, etc. The extremely realistic images (also known as AI-Generated Content, AIGC for short) produced by these models bring urgent needs for intellectual property protection such as data traceability and copyright certification. An attacker can steal the output images of the target model and use them as part of the training data to train a private surrogate UIG model. The implementation mechanisms of UIG models are diverse and complex, and there is no unified and effective protection and verification method at present. To address these issues, we propose a two-stage unified watermark verification mechanism with high-binding effects for such models. In the first stage, we use an encoder to invisibly write the watermark image into the output images of the original AIGC tool, and reversely extract the watermark image through the corresponding decoder. In the second stage, we design the decoder fine-tuning process, and the fine-tuned decoder can make correct judgments on whether the suspicious model steals the original AIGC tool data. Experiments demonstrate our method can complete the verification work with almost zero false positive rate under the condition of only using the model output images. Moreover, the proposed method can achieve data steal verification across different types of UIG models, which further increases the practicality of the method.

cs.CV↗

arXiv2023

Experimental topological quantum computing with electric circuits

The key obstacle to the realization of a scalable quantum computer is overcoming environmental and control errors. Topological quantum computation has attracted great attention because it has emerged as one of the most promising approaches to solving these problems. Various theoretical schemes for building topological quantum computation have been proposed. However, experimental implementation has always been a great challenge because it has proved to be extremely difficult to create and manipulate topological qubits in real systems. Therefore, topological quantum computation has not been realized in experiments yet. Here, we report the first experimental realization of topological quantum computation with electric circuits. Based on our proposed new scheme with circuits, Majorana-like edge states are not only observed experimentally, but also T junctions are constructed for the braiding process. Furthermore, we demonstrate the feasibility of topological quantum computing through a set of one- and two-qubit unitary operations. Finally, our implementation of Grover's search algorithm demonstrates that topological quantum computation is ideally suited for such tasks.

cond-mat.mes-hall↗

arXiv2023

Experimental observation of exceptional bound states in a classical circuit network

Exceptional bound (EB) states represent an unique new class of robust bound states protected by the defectiveness of non-Hermitian exceptional points. Conceptually distinct from the more well-known topological states and non-Hermitian skin states, they were recently discovered as a novel source of negative entanglement entropy in the quantum entanglement context. Yet, EB states have been physically elusive, being originally interpreted as negative probability eigenstates of the propagator of non-Hermitian Fermi gases. In this work, we show that EB states are in fact far more ubiquitous, also arising robustly in broad classes of systems whether classical or quantum. This hinges crucially on a newly-discovered spectral flow that rigorously justifies the EB nature of small candidate lattice systems. As a highlight, we present their first experimental realization through an electrical circuit, where they manifest as prominent stable resonant voltage profiles. Our work brings a hitherto elusive but fundamentally distinctive quantum phenomenon into the realm of classical metamaterials, and provides a novel pathway for the engineering of robust modes in otherwise sensitive systems.

quant-ph↗

arXiv2023

Designing a braille reader using the snap buckling of bistable magnetic shells

A design concept is introduced for the building block, a dot, of programmable braille readers utilizing bistable shell buckling, magnetic actuation, and pneumatic loading. The design process is guided by Finite Element simulations, which are initially validated through precision experiments conducted on a scaled-up, single-shell model system. Then, the simulations are leveraged to systematically explore the design space, adhering to the standardized geometric and physical specifications of braille systems. The findings demonstrate the feasibility of selecting design parameters that satisfy both geometric requirements and blocking forces under moderate magnetic fields, facilitated by pneumatic loading to switch between the two stable states. The advantages of the proposed design include the reversible bistability of the actuators and fast state-switching via a transient magnetic field. While the study is focused on experimentally validated numerical simulations, several manufacturing challenges that need to be resolved for future physical implementations are identified.

cond-mat.soft↗

arXiv2023

A maximum principle for progressive optimal control of mean-filed forward-backward stochastic system involving random jumps and impulse controls

In this paper, we study an optimal control problem of a mean-field forward-backward stochastic system with random jumps in progressive structure, where both regular and singular controls are considered in our formula. In virtue of the variational technology, the related stochastic maximum principle (SMP) has been obtained, and it is essentially different from that in the classical predictable structure. Specifically, there are three parts in our SMP, i.e. continuous part, jump part and impulse part, and they are respectively used to characterize the characteristics of the optimal controls at continuous time, jump time and impulse time. This shows that the progressive structure can more accurately describe the characteristics of the optimal control at the jump time. We also give two linear-quadratic (LQ) examples to show the significance of our results.

math.OC↗

arXiv2023

Performance Characterization of using Quantization for DNN Inference on Edge Devices: Extended Version

Quantization is a popular technique used in Deep Neural Networks (DNN) inference to reduce the size of models and improve the overall numerical performance by exploiting native hardware. This paper attempts to conduct an elaborate performance characterization of the benefits of using quantization techniques -- mainly FP16/INT8 variants with static and dynamic schemes -- using the MLPerf Edge Inference benchmarking methodology. The study is conducted on Intel x86 processors and Raspberry Pi device with ARM processor. The paper uses a number of DNN inference frameworks, including OpenVINO (for Intel CPUs only), TensorFlow Lite (TFLite), ONNX, and PyTorch with MobileNetV2, VGG-19, and DenseNet-121. The single-stream, multi-stream, and offline scenarios of the MLPerf Edge Inference benchmarks are used for measuring latency and throughput in our experiments. Our evaluation reveals that OpenVINO and TFLite are the most optimized frameworks for Intel CPUs and Raspberry Pi device, respectively. We observe no loss in accuracy except for the static quantization techniques. We also observed the benefits of using quantization for these optimized frameworks. For example, INT8-based quantized models deliver $3.3\times$ and $4\times$ better performance over FP32 using OpenVINO on Intel CPU and TFLite on Raspberry Pi device, respectively, for the MLPerf offline scenario. To the best of our knowledge, this paper is the first one that presents a unique characterization study characterizing the impact of quantization for a range of DNN inference frameworks -- including OpenVINO, TFLite, PyTorch, and ONNX -- on Intel x86 processors and Raspberry Pi device with ARM processor using the MLPerf Edge Inference benchmark methodology.

cs.PF↗

arXiv2022

Topolectrical-circuit realization of quadrupolar surface semimetals

Recent studies have shown that there are two types of topological quadrupolar semimetals. One is the quadrupolar bulk semimetal, which is a conventional semimetal with gapless nodes in the bulk spectrum. The other type is named the quadrupolar surface semimetal, possessing a gapped bulk but gapless nodes on the surface spectrum. There have been many experiments on the implementation of quadrupolar bulk semimetals. However, the quadrupolar surface semimetal has never been realized. Here we report on the experimental realization of the quadrupolar surface semimetal in the circuit system. To highlight the properties of the quadrupolar surface semimetal, we also prepare a quadrupolar bulk semimetal in the circuit as a contrast. It has been demonstrated that the quadrupolar surface semimetal has a bulk band gap, but two gapless nodes appear on the surfaces. Also, the electrons not only concentrate on the hinges but also on the corresponding surfaces. Our studies provide novel ways to control electrical signals and may have potential applications in the field of intergraded circuit design.

cond-mat.mes-hall↗

arXiv2022

Exploring topological phase transition andWeyl physics in five dimensions with electric circuits

Weyl semimetals are phases of matter with gapless electronic excitations that are protected by topology and symmetry. Their properties depend on the dimensions of the systems. It has been theoretically demonstrated that five-dimensional (5D) Weyl semimetals emerge as novel phases during the topological phase transition in analogy to the three-dimensional case. However, experimental observation of such a phenomenon remains a great challenge because the tunable 5D system is extremely hard to construct in real space. Here, we construct 5D electric circuit platforms in fully real space and experimentally observe topological phase transitions in five dimensions. Not only are Yang monopoles and linked Weyl surfaces observed experimentally, but various phase transitions in five dimensions are also proved, such as the phase transitions from a normal insulator to a Hopf link of twoWeyl surfaces and then to a 5D topological insulator. The demonstrated topological phase transitions in five dimensions leverage the concept of higher-dimensional Weyl physics to control electrical signals in the engineered circuits.

cond-mat.mes-hall↗

arXiv2022

On Semiparametric Efficiency of an Emerging Class of Regression Models for Between-subject Attributes

The semiparametric regression models have attracted increasing attention owing to their robustness compared to their parametric counterparts. This paper discusses the efficiency bound for functional response models (FRM), an emerging class of semiparametric regression that serves as a timely solution for research questions involving pairwise observations. This new paradigm is especially appealing to reduce astronomical data dimensions for those arising from wearable devices and high-throughput technology, such as microbiome Beta-diversity, viral genetic linkage, single-cell RNA sequencing, etc. Despite the growing applications, the efficiency of their estimators has not been investigated carefully due to the extreme difficulty to address the inherent correlations among pairs. Leveraging the Hilbert-space-based semiparametric efficiency theory for classical within-subject attributes, this manuscript extends such asymptotic efficiency into the broader regression involving between-subject attributes and pinpoints the most efficient estimator, which leads to a sensitive signal-detection in practice. With pairwise outcomes burgeoning immensely as effective dimension-reduction summaries, the established theory will not only fill the critical gap in identifying the most efficient semiparametric estimator but also propel wide-ranging implementations of this new paradigm for between-subject attributes.

stat.ME↗

arXiv2022

Continuous-time mean-variance portfolio selection under non-Markovian regime-switching model with random horizon

In this paper, we consider a continuous-time mean-variance portfolio selection with regime-switching and random horizon. Unlike previous works, the dynamic of assets are described by non-Markovian regime-switching models in the sense that all the market parameters are predictable with respect to the filtration generated jointly by Markov chain and Brownian motion. We formulate this problem as a constrained stochastic linear-quadratic optimal control problem. The Markov chain is assumed to be independent of the Brownian motion. So the market is incomplete. We derive closed-form expressions for both the optimal portfolios and the efficient frontier. All the results are different from those in the problem with fixed time horizon.

q-fin.MF↗