Not only cheaper, but (since in this case money ≈ hardware-cost × time), faster. They claim that training time can even approach inference time:
> EGGROLL's efficiency results in a hundredfold increase in training throughput for billion-parameter models at large population sizes, nearly reaching the throughput of pure batch inference
Their technique does not claim to compete with gradient descent - it's competition for techniques like Proximal Policy Optimization, so it's more suited for things like creating a reasoning model out of an existing pre-trained model.
Not only cheaper, but (since in this case money ≈ hardware-cost × time), faster. They claim that training time can even approach inference time:
> EGGROLL's efficiency results in a hundredfold increase in training throughput for billion-parameter models at large population sizes, nearly reaching the throughput of pure batch inference