AggregaThor
Framework over TensorFlow implementing robust stochastic gradient descent
Framework built over TensorFlow implementing state-of-the-art Byzantine-resilient, distributed Stochastic Gradient Descent (SGD). Modular approach allows most of its components to be reused in other projects. Unreliable communication channels are supported, providing a performance speed-up over standard TensorFlow in saturated networks.
inactive
—
entered showcase: 2019-03-18
—
entry updated: 2024-03-22
This project has not yet been evaluated by the C4DT Factory team.
We will be happy to evaluate it upon request.
Library, Application
Python / C++