Mingchen Li
YOU?
Author Swipe
View article: MS-DFTVNet:A Long-Term Time Series Prediction Method Based on Multi-Scale Deformable Convolution
MS-DFTVNet:A Long-Term Time Series Prediction Method Based on Multi-Scale Deformable Convolution Open
Research on long-term time series prediction has primarily relied on Transformer and MLP models, while the potential of convolutional networks in this domain remains underexplored. To address this, we propose a novel multi-scale time serie…
View article: TVNet: A Novel Time Series Analysis Method Based on Dynamic Convolution and 3D-Variation
TVNet: A Novel Time Series Analysis Method Based on Dynamic Convolution and 3D-Variation Open
With the recent development and advancement of Transformer and MLP architectures, significant strides have been made in time series analysis. Conversely, the performance of Convolutional Neural Networks (CNNs) in time series analysis has f…
View article: Spatial Metabolomics Combined with MALDI-MSI Unveils Gut-Brain Axis Mechanisms of Angelica dahurica Radix in Migraine Rats
Spatial Metabolomics Combined with MALDI-MSI Unveils Gut-Brain Axis Mechanisms of Angelica dahurica Radix in Migraine Rats Open
View article: PETA: evaluating the impact of protein transfer learning with sub-word tokenization on downstream applications
PETA: evaluating the impact of protein transfer learning with sub-word tokenization on downstream applications Open
View article: Simple, Efficient and Scalable Structure-aware Adapter Boosts Protein Language Models
Simple, Efficient and Scalable Structure-aware Adapter Boosts Protein Language Models Open
Fine-tuning Pre-trained protein language models (PLMs) has emerged as a prominent strategy for enhancing downstream prediction tasks, often outperforming traditional supervised learning approaches. As a widely applied powerful technique in…
View article: Class-Attribute Priors: Adapting Optimization to Heterogeneity and Fairness Objective
Class-Attribute Priors: Adapting Optimization to Heterogeneity and Fairness Objective Open
Modern classification problems exhibit heterogeneities across individual classes: Each class may have unique attributes, such as sample size, label quality, or predictability (easy vs difficult), and variable importance at test-time. Witho…
View article: An interval constraint-based trading strategy with social sentiment for the stock market
An interval constraint-based trading strategy with social sentiment for the stock market Open
Developing effective strategies to earn excess returns in the stock market is a cutting-edge topic in the field of economics. At the same time, stock price forecasting that supports trading strategies is considered one of the most challeng…
View article: Enhancing the efficiency of protein language models with minimal wet-lab data through few-shot learning
Enhancing the efficiency of protein language models with minimal wet-lab data through few-shot learning Open
Accurately modeling the protein fitness landscapes holds great importance for protein engineering. Recently, due to their capacity and representation ability, pre-trained protein language models have achieved state-of-the-art performance i…
View article: PETA: Evaluating the Impact of Protein Transfer Learning with Sub-word Tokenization on Downstream Applications
PETA: Evaluating the Impact of Protein Transfer Learning with Sub-word Tokenization on Downstream Applications Open
Large protein language models are adept at capturing the underlying evolutionary information in primary structures, offering significant practical value for protein engineering. Compared to natural language models, protein amino acid seque…
View article: Can multi-source heterogeneous data improve the forecasting performance of tourist arrivals amid COVID-19? Mixed-data sampling approach
Can multi-source heterogeneous data improve the forecasting performance of tourist arrivals amid COVID-19? Mixed-data sampling approach Open
View article: SESNet: sequence-structure feature-integrated deep learning method for data-efficient protein engineering
SESNet: sequence-structure feature-integrated deep learning method for data-efficient protein engineering Open
View article: SESNet: sequence-structure feature-integrated deep learning method for data-efficient protein engineering
SESNet: sequence-structure feature-integrated deep learning method for data-efficient protein engineering Open
Deep learning has been widely used for protein engineering. However, it is limited by the lack of sufficient experimental data to train an accurate model for predicting the functional fitness of high-order mutants. Here, we develop SESNet,…
View article: What Can Be Learned from the Historical Trend of Crude Oil Prices? An Ensemble Approach to Crude Oil Price Forecasting
What Can Be Learned from the Historical Trend of Crude Oil Prices? An Ensemble Approach to Crude Oil Price Forecasting Open
View article: Nesfatin-1 protects H9c2 cardiomyocytes against cobalt chloride-induced hypoxic injury by modulating the MAPK and Notch1 signaling pathways
Nesfatin-1 protects H9c2 cardiomyocytes against cobalt chloride-induced hypoxic injury by modulating the MAPK and Notch1 signaling pathways Open
View article: Multiband Triple L-Arms Patch Antenna with Diamond Slot Ground for 5G Applications
Multiband Triple L-Arms Patch Antenna with Diamond Slot Ground for 5G Applications Open
This paper reported a pioneering 5G multiband microstrip line fed patch antenna for IoT, wireless power transfer (WPT) and data transmission. The proposed antenna is accomplished using a triple L-arms patch antenna responsible for the mult…
View article: Generalization Guarantees for Neural Networks via Harnessing the Low-rank Structure of the Jacobian
Generalization Guarantees for Neural Networks via Harnessing the Low-rank Structure of the Jacobian Open
Modern neural network architectures often generalize well despite containing many more parameters than the size of the training dataset. This paper explores the generalization capabilities of neural networks trained via gradient descent. W…