Long Short Term Memory
Long Short-Term Memory (LSTM) networks are a type of recurrent neural network designed to process sequential data by learning long-term dependencies, enabling accurate predictions and classifications in various applications. Current research focuses on enhancing LSTM architectures, such as incorporating convolutional layers, attention mechanisms, and hybrid models combining LSTMs with other deep learning techniques like transformers or graph neural networks, to improve efficiency and accuracy. This work is significant because LSTMs are proving highly effective across diverse fields, from financial forecasting and environmental monitoring to medical image analysis and activity recognition, offering powerful tools for analyzing complex temporal data.
Papers
A comparative study of statistical and machine learning models on near-real-time daily emissions prediction
Xiangqian Li
A Light-weight CNN Model for Efficient Parkinson's Disease Diagnostics
Xuechao Wang, Junqing Huang, Marianna Chatzakou, Kadri Medijainen, Pille Taba, Aaro Toomela, Sven Nomm, Michael Ruzhansky