Shapeshifter: a parameter-efficient transformer using factorized reshaped matrices

A Panahi, S Saeedi, T Arodz - Advances in Neural …, 2021 - proceedings.neurips.cc
Abstract Language models employ a very large number of trainable parameters. Despite
being highly overparameterized, these networks often achieve good out-of-sample test …

TT-GNN: Efficient On-Chip Graph Neural Network Training via Embedding Reformation and Hardware Optimization

Z Qu, D Niu, S Li, H Zheng, Y ** for Tensorized Neural Network Processing
JF Zhang, CH Lu, Z Zhang - IEEE Transactions on Computers, 2024 - ieeexplore.ieee.org
The continuous growth of deep neural network model size and complexity hinders the
adoption of large models in resource-constrained platforms. Tensor decomposition has …

Tensor Network-Based Lightweight Energy Forecasting for Virtual Power Plant

D Watari, H Tanimoto, T Okubo, S Todo… - 2024 13th …, 2024 - ieeexplore.ieee.org
Virtual power plant (VPP) operations demand lightweight energy forecasting models due to
increasing computational requirements. This paper applies tensor-train (TT) decomposition …

[BOOK][B] Addressing Data Explosion Issue in Emerging Deep Learning Applications

Z Qu - 2023 - search.proquest.com
With the continuous booming development of deep learning, many kinds of model variants
are being proposed to tackle more difficult machine learning tasks, such as Transformers …

Multivariate Analysis and Machine Learning towards Security Applications

Y Shin - 2022 - search.proquest.com
Analyzing the high-dimensional dataset is crucial for recent security applications. However,
detecting anomalies from practical physical systems is a challenging task due to their huge …