Migel D. Tissera
Papers
1
Total Citations
7
H-Index
1
About
Migel D. Tissera is a researcher whose work lies at the intersection of neural network architecture and efficient machine learning, with a particular focus on developing algorithms that reduce computational overhead without sacrificing performance. His most notable contribution, the 2016 paper "Modular expansion of the hidden layer in Single Layer Feedforward Neural Networks," introduces a novel training algorithm that enables rapid learning through a modular, constructive expansion of the output weights layer. This approach is designed to operate with minimal computational processing power, memory, and time—a critical advancement for deploying neural networks on resource-constrained devices. With 7 citations, this work has garnered attention for its practical implications in real-time and embedded systems. Tissera’s research addresses a key bottleneck in deep learning: the trade-off between model complexity and efficiency. By proposing a scalable, low-cost training method, he has contributed to making neural networks more accessible for applications where speed and resource economy are paramount. His work continues to inspire further exploration into modular and adaptive network designs.
Research Focus
Key Achievements
Top Papers
- 1