Abstract: We investigated the adaptation and performance of Masked Autoencoders (MAEs) with Vision Transformer (ViT) architectures for self-supervised representation learning on one-dimensional (1D) ...

Abstract: Denoising images is widely used in applications from critical medical systems to software based image enhancement in our cell phones. The natural noise is simulated by adding noise to an ...

Deep Autoencoder for Handwritten Digits MNIST This project demonstrates the implementation of a Deep Autoencoder using Keras and TensorFlow to encode and reconstruct handwritten digits from the ...

Variational Autoencoders (VAEs) are an artificial neural network architecture to generate new data. They are similar to regular autoencoders, which consist of an encoder and decoder. The encoder takes ...

A research project investigating how Reinforcement Learning from Human Feedback (RLHF) systematically reshapes cultural marker representations in language models using Sparse Autoencoders (SAEs) with ...

ABSTRACT: Video-based anomaly detection in urban surveillance faces a fundamental challenge: scale-projective ambiguity. This occurs when objects of different physical sizes appear identical in camera ...