Unbiased scalable softmax optimization
This addresses the computational bottleneck in training neural networks and language models with many categories, providing an unbiased scalable solution.
The paper tackled the problem of efficiently computing unbiased estimates for softmax likelihood in large-scale models, achieving comprehensive outperformance over state-of-the-art methods on seven real-world datasets.
Recent neural network and language models rely on softmax distributions with an extremely large number of categories. Since calculating the softmax normalizing constant in this context is prohibitively expensive, there is a growing literature of efficiently computable but biased estimates of the softmax. In this paper we propose the first unbiased algorithms for maximizing the softmax likelihood whose work per iteration is independent of the number of classes and datapoints (and no extra work is required at the end of each epoch). We show that our proposed unbiased methods comprehensively outperform the state-of-the-art on seven real world datasets.