Ali Basirat's Projects
Boltzmann Machines in TensorFlow with examples
Differentiable Perturb-and-Parse operator
A plug-and-play adapter architecture that efficiently adapts large language models to downstream tasks. Essential for client-server architecture language model services. A practical approach to adapting LLMs' hidden activations without modifying the base model.
We use principal component analysis for word embedding. The method is able to process both annotated and raw corpora.
An extention of the randomized singular value decomposition (SVD) algorithm to estimate the SVD of a shifted data matrix without explicitly constructing the matrix in the memory. With no loss in the accuracy of the original algorithm, the extended algorithm provides for a more efficient way of matrix factorization. The algorithm facilitates the low-rank approximation and principal component analysis (PCA) of off-center data matrices. When applied to different types of data matrices, our experimental results confirm the advantages of the extensions made to the original algorithm.
A multi-task neural network model for multi-tagging of natural languages
A transition-based parser for Universal Dependencies with BiLSTM word and character representations.