arXiv subjects

Siyu Liu

Publications and source records attributed to Siyu Liu.

At least 55 records · Page 3Linked to original sources

arXiv2024

Large Language Models for Material Property Predictions: elastic constant tensor prediction and materials design

Efficient and accurate prediction of material properties is critical for advancing materials design and applications. The rapid-evolution of large language models (LLMs) presents a new opportunity for material property predictions, complementing experimental measurements and multi-scale computational methods. We focus on predicting the elastic constant tensor, as a case study, and develop domain-specific LLMs for predicting elastic constants and for materials discovery. The proposed ElaTBot LLM enables simultaneous prediction of elastic constant tensors, bulk modulus at finite temperatures, and the generation of new materials with targeted properties. Moreover, the capabilities of ElaTBot are further enhanced by integrating with general LLMs (GPT-4o) and Retrieval-Augmented Generation (RAG) for prediction. A specialized variant, ElaTBot-DFT, designed for 0 K elastic constant tensor prediction, reduces the prediction errors by 33.1% compared with domain-specific, material science LLMs (Darwin) trained on the same dataset. This natural language-based approach lowers the barriers to computational materials science and highlights the broader potential of LLMs for material property predictions and inverse design.

cond-mat.mtrl-sci↗

arXiv2024

IterSelectTune: An Iterative Training Framework for Efficient Instruction-Tuning Data Selection

As large language models (LLMs) continue to advance, instruction tuning has become critical for improving their ability to generate accurate and contextually appropriate responses. Although numerous instruction-tuning datasets have been developed to enhance LLM performance, selecting high-quality instruction data from large source datasets typically demands significant human effort. In this work, we introduce $\textbf{IterSelectTune}$, an efficient, cost-effective iterative training policy for selecting high-quality instruction data with no human involvement and limited reliance on GPT-4. By fine-tuning on approximately 20\% of the source data, our method consistently outperforms models fine-tuned on the full dataset across multiple benchmarks and public test datasets. These results highlight the effectiveness of our approach in enhancing LLM performance while reducing the computational resources required for instruction tuning.

cs.CL↗

arXiv2024

Additive mappings preserving orthogonality between complex inner product spaces

Let $H$ and $K$ be two complex inner product spaces with dim$(X)\geq 2$. We prove that for each non-zero additive mapping $A:H \to K$ with dense image the following statements are equivalent: $(a)$ $A$ is (complex) linear or conjugate-linear mapping and there exists $γ>0$ such that $\| A (x) \| = γ\|x\|$, for all $x\in X$, that is, $A$ is a positive scalar multiple of a linear or a conjugate-linear isometry; $(b)$ There exists $γ_1 >0$ such that one of the next properties holds for all $x,y \in H$: $(b.1)$ $\langle A(x) |A(y)\rangle = γ_1 \langle x|y\rangle,$ $(b.2)$ $\langle A(x) |A(y)\rangle = γ_1 \langle y|x \rangle;$ $(c)$ $A$ is linear or conjugate-linear and preserves orthogonality in both directions; $(d)$ $A$ is linear or conjugate-linear and preserves orthogonality; $(e)$ $A$ is additive and preserves orthogonality in both directions; $(f)$ $A$ is additive and preserves orthogonality. This extends to the complex setting a recent generalization of the Koldobsky--Blanco--Turnšek theorem obtained by Wójcik for real normed spaces.

math.FA↗

arXiv2024

Microwave Photonic Multi-Mode Injection-Locked Frequency Divider With a Wide Operational Range Based on an Optoelectronic Oscillator

We propose and implement a microwave photonic multi-mode injection-locked frequency divider (ILFD) with a wide frequency operational range based on an optoelectronic oscillator (OEO). In the OEO, a Mach-Zehnder modulator (MZM) and a photodetector (PD) are employed to construct a frequency multiplier to achieve an N-1 times frequency multiplication, which is then mixed with an external injection signal at an electrical mixer in the OEO loop. By adjusting the round-trip gain and time delay of the OEO loop, a radio frequency (RF) signal with a frequency that is 1/N that of the injection signal is generated, thus N times frequency division is achieved. Theoretical analysis and experimental verification are conducted to evaluate the effectiveness of the proposed ILFD. The results demonstrate that the system can divide a RF signal from 2.6 to 20.8 GHz to 1.3 to 1.95 GHz with different frequency division factors ranging from 2 to 13. A significant improvement in phase noise of 35.11 dB is also obtained at a frequency offset of 100 kHz when the frequency division factor is 13.

physics.optics↗

arXiv2024

RetAssist: Facilitating Vocabulary Learners with Generative Images in Story Retelling Practices

Reading and repeatedly retelling a short story is a common and effective approach to learning the meanings and usages of target words. However, learners often struggle with comprehending, recalling, and retelling the story contexts of these target words. Inspired by the Cognitive Theory of Multimedia Learning, we propose a computational workflow to generate relevant images paired with stories. Based on the workflow, we work with learners and teachers to iteratively design an interactive vocabulary learning system named RetAssist. It can generate sentence-level images of a story to facilitate the understanding and recall of the target words in the story retelling practices. Our within-subjects study (N=24) shows that compared to a baseline system without generative images, RetAssist significantly improves learners' fluency in expressing with target words. Participants also feel that RetAssist eases their learning workload and is more useful. We discuss insights into leveraging text-to-image generative models to support learning tasks.

cs.HC↗

arXiv2024

A Prompt-Engineered Large Language Model, Deep Learning Workflow for Materials Classification

Large language models (LLMs) have demonstrated rapid progress across a wide array of domains. Owing to the very large number of parameters and training data in LLMs, these models inherently encompass an expansive and comprehensive materials knowledge database, far exceeding the capabilities of individual researcher. Nonetheless, devising methods to harness the knowledge embedded within LLMs for the design and discovery of novel materials remains a formidable challenge. We introduce a general approach for addressing materials classification problems, which incorporates LLMs, prompt engineering, and deep learning. Utilizing a dataset of metallic glasses as a case study, our methodology achieved an improvement of up to 463% in prediction accuracy compared to conventional classification models. These findings underscore the potential of leveraging textual knowledge generated by LLMs for materials especially in the common situation where datasets are sparse, thereby promoting innovation in materials discovery and design.

cond-mat.mtrl-sci↗

arXiv2024

A comprehensive survey on deep active learning in medical image analysis

Deep learning has achieved widespread success in medical image analysis, leading to an increasing demand for large-scale expert-annotated medical image datasets. Yet, the high cost of annotating medical images severely hampers the development of deep learning in this field. To reduce annotation costs, active learning aims to select the most informative samples for annotation and train high-performance models with as few labeled samples as possible. In this survey, we review the core methods of active learning, including the evaluation of informativeness and sampling strategy. For the first time, we provide a detailed summary of the integration of active learning with other label-efficient techniques, such as semi-supervised, self-supervised learning, and so on. We also summarize active learning works that are specifically tailored to medical image analysis. Additionally, we conduct a thorough comparative analysis of the performance of different AL methods in medical image analysis with experiments. In the end, we offer our perspectives on the future trends and challenges of active learning and its applications in medical image analysis.

cs.CV↗

arXiv2023

Constraining the Woods-Saxon potential in fusion reactions based on the neural network

The accurate determination of the nuclear interaction potential is essential for predicting the fusion cross sections and understanding the reaction mechanism, which plays an important role in the synthesis of superheavy elements. In this work, the neural network, which combines with the calculations of the fusion cross sections via the Hill-Wheeler formula, is developed to optimize the parameters of the Woods-Saxon potential by comparing the experimental values. The correlations between the parameters of Woods-Saxon potential and the reaction partners, which can be quantitatively fitted to a sigmoid-like function with the mass numbers, have been displayed manifestly for the first time. This study could promote the accurate estimation of nucleus-nucleus interaction potential in low energy heavy-ion collisions.

nucl-th↗

arXiv2023

Cascaded Multi-Modal Mixing Transformers for Alzheimer's Disease Classification with Incomplete Data

Accurate medical classification requires a large number of multi-modal data, and in many cases, different feature types. Previous studies have shown promising results when using multi-modal data, outperforming single-modality models when classifying diseases such as Alzheimer's Disease (AD). However, those models are usually not flexible enough to handle missing modalities. Currently, the most common workaround is discarding samples with missing modalities which leads to considerable data under-utilization. Adding to the fact that labeled medical images are already scarce, the performance of data-driven methods like deep learning can be severely hampered. Therefore, a multi-modal method that can handle missing data in various clinical settings is highly desirable. In this paper, we present Multi-Modal Mixing Transformer (3MAT), a disease classification transformer that not only leverages multi-modal data but also handles missing data scenarios. In this work, we test 3MT for AD and Cognitively normal (CN) classification and mild cognitive impairment (MCI) conversion prediction to progressive MCI (pMCI) or stable MCI (sMCI) using clinical and neuroimaging data. The model uses a novel Cascaded Modality Transformer architecture with cross-attention to incorporate multi-modal information for more informed predictions. We propose a novel modality dropout mechanism to ensure an unprecedented level of modality independence and robustness to handle missing data scenarios. The result is a versatile network that enables the mixing of arbitrary numbers of modalities with different feature types and also ensures full data utilization missing data scenarios. The model is trained and evaluated on the ADNI dataset with the SOTRA performance and further evaluated with the AIBL dataset with missing data.

eess.IV↗

arXiv2023

TriFormer: A Multi-modal Transformer Framework For Mild Cognitive Impairment Conversion Prediction

The prediction of mild cognitive impairment (MCI) conversion to Alzheimer's disease (AD) is important for early treatment to prevent or slow the progression of AD. To accurately predict the MCI conversion to stable MCI or progressive MCI, we propose Triformer, a novel transformer-based framework with three specialized transformers to incorporate multi-model data. Triformer uses I) an image transformer to extract multi-view image features from medical scans, II) a clinical transformer to embed and correlate multi-modal clinical data, and III) a modality fusion transformer that produces an accurate prediction based on fusing the outputs from the image and clinical transformers. Triformer is evaluated on the Alzheimer's Disease Neuroimaging Initiative (ANDI)1 and ADNI2 datasets and outperforms previous state-of-the-art single and multi-modal methods.

cs.CV↗

arXiv2023

State estimation of a carbon capture process through POD model reduction and neural network approximation

This paper presents an efficient approach for state estimation of post-combustion CO2 capture plants (PCCPs) by using reduced-order neural network models. The method involves extracting lower-dimensional feature vectors from high-dimensional operational data of the PCCP and constructing a reduced-order process model using proper orthogonal decomposition (POD). Multi-layer perceptron (MLP) neural networks capture the dominant dynamics of the process and train the network parameters with low-dimensional data obtained from open-loop simulations. The proposed POD-MLP model can be used as the basis for estimating the states of PCCPs at a significantly decreased computational cost. For state estimation, a reduced-order extended Kalman filtering (EKF) scheme based on the POD-MLP model is developed. Our simulations demonstrate that the proposed POD-MLP modeling approach reduces computational complexity compared to the POD-only model for nonlinear systems. Additionally, the POD-MLP-EKF algorithm can accurately reconstruct the full state information of PCCPs while significantly improving computational efficiency compared to the EKF based on the original PCCP model.

eess.SY↗

arXiv2023

Sensor network design for post-combustion CO2 capture plants: economy, complexity and robustness

State estimation is crucial for the monitoring and control of post-combustion CO2 capture plants (PCCPs). The performance of state estimation is highly reliant on the configuration of sensors. In this work, we consider the problem of sensor selection for PCCPs and propose a computationally efficient method to determine an appropriate number of sensors and the corresponding placement of the sensors. The objective is to find the (near-)optimal set of sensors that provides the maximum degree of observability for state estimation while satisfying the budget constraint. Specifically, we resort to the information contained in the sensitivity matrix calculated around the operating region of a PCCP to quantify the degree of observability of the entire system corresponding to the placed sensors. The sensor selection problem is converted to an optimization problem, and is efficiently solved by a one-by-one removal approach through sensitivity analysis. Next, we extend our approach to study fault tolerance (resilience) of the selected sensors to sensor malfunction. The resilient sensor selection problem is to find a sensor network that gives good estimation performance even when some of the sensors fail, thereby improving the overall system robustness. The resilient sensor selection problem is formulated as a max-min optimization problem. We show how the proposed approach can be adapted to solve the sensor selection max-min optimization problem. By implementing the proposed approaches, the sensor network is configured for the PCCP efficiently.

eess.SY↗

arXiv2023

Explainable Semantic Medical Image Segmentation with Style

Semantic medical image segmentation using deep learning has recently achieved high accuracy, making it appealing to clinical problems such as radiation therapy. However, the lack of high-quality semantically labelled data remains a challenge leading to model brittleness to small shifts to input data. Most works require extra data for semi-supervised learning and lack the interpretability of the boundaries of the training data distribution during training, which is essential for model deployment in clinical practice. We propose a fully supervised generative framework that can achieve generalisable segmentation with only limited labelled data by simultaneously constructing an explorable manifold during training. The proposed approach creates medical image style paired with a segmentation task driven discriminator incorporating end-to-end adversarial training. The discriminator is generalised to small domain shifts as much as permissible by the training data, and the generator automatically diversifies the training samples using a manifold of input features learnt during segmentation. All the while, the discriminator guides the manifold learning by supervising the semantic content and fine-grained features separately during the image diversification. After training, visualisation of the learnt manifold from the generator is available to interpret the model limits. Experiments on a fully semantic, publicly available pelvis dataset demonstrated that our method is more generalisable to shifts than other state-of-the-art methods while being more explainable using an explorable manifold.

eess.IV↗

arXiv2022

Advisory Tool for Managing Failure Cascades in Systems with Wind Power

This paper concerns the resilience of systems with wind power upon wind reduction by evaluating the potential of corrective actions, such as generation and load dispatch, on minimizing the effects of transmission line failures. Three functions (grid, consumer-centric loss, and resilience impact) are used to statistically evaluate the criticality of initial contingent failures and wind reductions. Our model is learned with Monte Carlo, convex optimization, and adaptive selection, illustrated on the IEEE-30 and IEEE-300 bus systems with both AC and DC models. We highlight the impact of wind reductions and propose physically implementable solutions.

eess.SY↗

arXiv2022

Generalisable 3D Fabric Architecture for Streamlined Universal Multi-Dataset Medical Image Segmentation

Data scarcity is common in deep learning models for medical image segmentation. Previous works proposed multi-dataset learning, either simultaneously or via transfer learning to expand training sets. However, medical image datasets have diverse-sized images and features, and developing a model simultaneously for multiple datasets is challenging. This work proposes Fabric Image Representation Encoding Network (FIRENet), a universal architecture for simultaneous multi-dataset segmentation and transfer learning involving arbitrary numbers of dataset(s). To handle different-sized image and feature, a 3D fabric module is used to encapsulate many multi-scale sub-architectures. An optimal combination of these sub-architectures can be implicitly learnt to best suit the target dataset(s). For diverse-scale feature extraction, a 3D extension of atrous spatial pyramid pooling (ASPP3D) is used in each fabric node for a fine-grained coverage of rich-scale image features. In the first experiment, FIRENet performed 3D universal bone segmentation of multiple musculoskeletal datasets of the human knee, shoulder and hip joints and exhibited excellent simultaneous multi-dataset segmentation performance. When tested for transfer learning, FIRENet further exhibited excellent single dataset performance (when pre-training on a prostate dataset), as well as significantly improved universal bone segmentation performance. The following experiment involves the simultaneous segmentation of the 10 Medical Segmentation Decathlon (MSD) challenge datasets. FIRENet demonstrated good multi-dataset segmentation results and inter-dataset adaptability of highly diverse image sizes. In both experiments, FIRENet's streamlined multi-dataset learning with one unified network that requires no hyper-parameter tuning.

eess.IV↗

arXiv2022

Towards Statistical Methods for Minimizing Effects of Failure Cascades

This paper concerns the potential of corrective actions, such as generation and load dispatch on minimizing the effects of transmission line failures in electric power systems. Three loss functions (grid-centric, consumer-centric, and influence localization) are used to statistically evaluate the criticality of initial contingent failures. A learning scheme for both AC and DC grid models combine a Monte Carlo approach with a convex dynamic programming formulation and introduces an adaptive selection process, illustrated on the IEEE-30 bus system.

eess.SY↗

arXiv2022

Structure Guided Manifolds for Discovery of Disease Characteristics

In medical image analysis, the subtle visual characteristics of many diseases are challenging to discern, particularly due to the lack of paired data. For example, in mild Alzheimer's Disease (AD), brain tissue atrophy can be difficult to observe from pure imaging data, especially without paired AD and Cognitively Normal ( CN ) data for comparison. This work presents Disease Discovery GAN ( DiDiGAN), a weakly-supervised style-based framework for discovering and visualising subtle disease features. DiDiGAN learns a disease manifold of AD and CN visual characteristics, and the style codes sampled from this manifold are imposed onto an anatomical structural "blueprint" to synthesise paired AD and CN magnetic resonance images (MRIs). To suppress non-disease-related variations between the generated AD and CN pairs, DiDiGAN leverages a structural constraint with cycle consistency and anti-aliasing to enforce anatomical correspondence. When tested on the Alzheimer's Disease Neuroimaging Initiative ( ADNI) dataset, DiDiGAN showed key AD characteristics (reduced hippocampal volume, ventricular enlargement, and atrophy of cortical structures) through synthesising paired AD and CN scans. The qualitative results were backed up by automated brain volume analysis, where systematic pair-wise reductions in brain tissue structures were also measured

eess.IV↗

arXiv2022

Towards Label-efficient Automatic Diagnosis and Analysis: A Comprehensive Survey of Advanced Deep Learning-based Weakly-supervised, Semi-supervised and Self-supervised Techniques in Histopathological Image Analysis

Histopathological images contain abundant phenotypic information and pathological patterns, which are the gold standards for disease diagnosis and essential for the prediction of patient prognosis and treatment outcome. In recent years, computer-automated analysis techniques for histopathological images have been urgently required in clinical practice, and deep learning methods represented by convolutional neural networks have gradually become the mainstream in the field of digital pathology. However, obtaining large numbers of fine-grained annotated data in this field is a very expensive and difficult task, which hinders the further development of traditional supervised algorithms based on large numbers of annotated data. More recent studies have started to liberate from the traditional supervised paradigm, and the most representative ones are the studies on weakly supervised learning paradigm based on weak annotation, semi-supervised learning paradigm based on limited annotation, and self-supervised learning paradigm based on pathological image representation learning. These new methods have led a new wave of automatic pathological image diagnosis and analysis targeted at annotation efficiency. With a survey of over 130 papers, we present a comprehensive and systematic review of the latest studies on weakly supervised learning, semi-supervised learning, and self-supervised learning in the field of computational pathology from both technical and methodological perspectives. Finally, we present the key challenges and future trends for these techniques.

cs.CV↗