News

Frontiers27d
Integrating multi-source monitoring data and deep convolutional autoencoder technology for slope failure pattern recognition
Methods: This study integrates rainfall, surface displacement, and vertical displacement monitoring data, and proposes an automatic failure mode identification method based on deep convolutional ...
GitHub27d
“Decoding Dense Embeddings: Sparse Autoencoders for Interpreting and Discretizing Dense Retrieval.”
To train the SAE, you need embeddings generated by a DPR model. We use dense embeddings from SimLM for all 8.8M MSMARCO passages and train queries. After training the SAE, you can reconstruct the ...
IEEE28d
AlphaGrad: Normalized Gradient Descent for Adaptive Multi-loss Functions in EEG-based Motor Imagery Classification
AlphaGrad is the first method to automatically adjust multi-loss functions with differing metric scales, including mean square error, cross-entropy, and deep metric learning, within the context of ...
IEEE29d
Numerical Analysis of Sparse-Array Metasurface Antennas using Gaussian Basis Functions
Abstract: A sparse-array metasurface (MTS ... as well as the sheet impedance are expanded into a set of regularly spaced Gaussian basis functions. The sheet impedance expansion is carried out ...