Data Encoding for Byzantine-Resilient Distributed Gradient Descent

We consider distributed gradient computation, where both data and computation are distributed among m worker machines, t of which can be Byzantine adversaries, and a designated (master) node computes the model/parameter vector, iteratively using gradient descent (GD). The Byzantine adversaries can (collaboratively) deviate arbitrarily from their gradient computation. To solve this, we propose a method based on data encoding and (real) error correction to combat the adversarial behavior. We can tolerate up to t ≤ [m-1/2] corrupt worker nodes, which is information-theoretically optimal. Our method does not assume any probability distribution on the data. We develop a sparse encoding scheme which enables computationally efficient data encoding. We demonstrate a trade-off between the number of adversaries tolerated and the resource requirement (storage and computational complexity). As an example, our scheme incurs a constant overhead (storage and computational complexity) over that required by the distributed GD algorithm, without adversaries, for t ≤ m/3 .