Search

The prediction of disease risk factors can screen vulnerable groups for effective prevention and treatment, so as to reduce their morbidity and mortality. Machine learning has a great demand for high-quality labeling information, and labeling noise in medical big data poses a great challenge to efficient disease risk warning methods. Therefore, this project intends to study the robust learning algorithm and apply it to the early warning of infectious disease risk. A dynamic truncated loss model is proposed, which combines the traditional mutual entropy implicit weight feature with the mean variation feature. It is robust to label noise. A lower bound on training loss is constructed, and a method based on sampling rate is proposed to reduce the gradient of suspected samples to reduce the influence of noise on training results. The effectiveness of this method under different types of noise was verified by using a stroke screening data set as an example. This method enables robust learning of data containing label noise.

The research explores the utilization of a deep learning model employing an attention mechanism in medical text mining. It targets the challenge of analyzing unstructured text information within medical data. This research seeks to enhance the model's capability to identify essential medical information by incorporating deep learning and attention mechanisms. This paper reviews the basic principles and typical model architecture of attention mechanisms and shows the effectiveness of their application in the tasks of disease prediction, drug side effect monitoring, and entity relationship extraction. Aiming at the particularity of medical texts, an adaptive attention model integrating domain knowledge is proposed, and its ability to understand medical terms and process complex contexts is optimized. The experiment verifies the model's effectiveness in improving task accuracy and robustness, especially when dealing with long text. The future research path of enhancing model interpretation, realizing cross-domain knowledge transfer, and adapting to low-resource scenarios is discussed in the research outlook, which provides a new perspective and method support for intelligent medical information processing and clinical decision assistance. Finally, cross-domain knowledge transfer and adaptation strategies for low-resource scenarios, providing theoretical basis and technical reference for promoting the development of intelligent medical information processing and clinical decision support systems., Comment: arXiv admin note: text overlap with arXiv:2405.11704 by other authors

In this paper, a tool for detecting LLM AI text generation is developed based on the Transformer model, aiming to improve the accuracy of AI text generation detection and provide reference for subsequent research. Firstly the text is Unicode normalised, converted to lowercase form, characters other than non-alphabetic characters and punctuation marks are removed by regular expressions, spaces are added around punctuation marks, first and last spaces are removed, consecutive ellipses are replaced with single spaces and the text is connected using the specified delimiter. Next remove non-alphabetic characters and extra whitespace characters, replace multiple consecutive whitespace characters with a single space and again convert to lowercase form. The deep learning model combines layers such as LSTM, Transformer and CNN for text classification or sequence labelling tasks. The training and validation sets show that the model loss decreases from 0.127 to 0.005 and accuracy increases from 94.96 to 99.8, indicating that the model has good detection and classification ability for AI generated text. The test set confusion matrix and accuracy show that the model has 99% prediction accuracy for AI-generated text, with a precision of 0.99, a recall of 1, and an f1 score of 0.99, achieving a very high classification accuracy. Looking forward, it has the prospect of wide application in the field of AI text detection., Comment: 6 pages

The emergence of pretrained large language models has led to the deployment of a range of social chatbots for chitchat. Although these chatbots demonstrate language ability and fluency, they are not guaranteed to be engaging and can struggle to retain users. This work investigates the development of social chatbots that prioritize user engagement to enhance retention, specifically examining the use of human feedback to efficiently develop highly engaging chatbots. The proposed approach uses automatic pseudo-labels collected from user interactions to train a reward model that can be used to reject low-scoring sample responses generated by the chatbot model at inference time. Intuitive evaluation metrics, such as mean conversation length (MCL), are introduced as proxies to measure the level of engagement of deployed chatbots. A/B testing on groups of 10,000 new daily chatbot users on the Chai Research platform shows that this approach increases the MCL by up to 70%, which translates to a more than 30% increase in user retention for a GPT-J 6B model. Future work aims to use the reward model to realise a data fly-wheel, where the latest user conversations can be used to alternately fine-tune the language model and the reward model.

Free-space optical communication is a promising means to establish versatile, secure and high-bandwidth communication for many critical point-to-point applications. While the spatial modes of light offer an additional degree of freedom to increase the information capacity of an optical link, atmospheric turbulence can introduce severe distortion to the spatial modes and lead to data degradation. Here, we propose and demonstrate a vector-beam-based, turbulence-resilient communication protocol, namely spatial polarization differential phase shift keying (SPDPSK), that can encode a large number of information levels using orthogonal spatial polarization states of light. We show experimentally that the spatial polarization profiles of the vector modes are resilient to atmospheric turbulence, and therefore can reliably transmit high-dimensional information through a turbid channel without the need of any adaptive optics for beam compensation. We construct a proof-of-principle experiment with a controllable turbulence cell. Using 34 vector modes, we have measured a channel capacity of 4.84 bits per pulse (corresponding to a data error rate of 4.3\%) through a turbulent channel with a scintillation index larger than 1. Our SPDPSK protocol can also effectively transmit 4.02 bits of information per pulse using 18 vector modes through even stronger turbulence with a scintillation index of 1.54. Our study provides direct experimental evidence on how the spatial polarization profiles of vector beams are resilient to atmospheric turbulence and paves the way towards practical, high-capacity, free-space communication solutions with robust performance under harsh turbulent environments.

The Hermite-Gaussian (HG) modes, sometimes also referred to as transverse electromagnetic modes in free space, form a complete and orthonormal basis that have been extensively used to describe optical fields. In addition, these modes have been shown to be helpful to enhance information capacity of optical communications as well as to achieve super-resolution imaging in microscopy. Here we propose and present the realization of an efficient, robust mode sorter that can sort a large number of HG modes based on the relation between HG modes and Laguerre-Gaussian (LG) modes. We experimentally demonstrate the sorting of 16 HG modes, and our method can be readily extended to a higher-dimensional state space in a straightforward manner. We expect that our demonstration will have direct applications in a variety of fields including fiber optics, classical and quantum communications, as well as super-resolution imaging.