Machine Learning

3 readers

1 users here now

Machine learning (ML) is a field devoted to understanding and building methods that let machines "learn" – that is, methods that leverage data to improve computer performance on some set of tasks. Machine learning algorithms build a model based on sample data, known as training data, in order to make predictions or decisions without being explicitly programmed to do so. Machine learning algorithms are used in a wide variety of applications, such as in medicine, email filtering, speech recognition, agriculture, and computer vision, where it is difficult or unfeasible to develop conventional algorithms to perform the needed tasks.

founded 1 year ago

Sophia: A Scalable Stochastic Second-order Optimizer for Language Model Pre-training (arxiv.org)

submitted 1 year ago by nsa@kbin.social to c/machinelearning@kbin.social

1 comments fedilink hide all child comments

Abstract:

Given the massive cost of language model pre-training, a non-trivial improvement of the optimization algorithm would lead to a material reduction on the time and cost of training. Adam and its variants have been state-of-the-art for years, and more sophisticated second-order (Hessian-based) optimizers often incur too much per-step overhead. In this paper, we propose Sophia, Second-order Clipped Stochastic Optimization, a simple scalable second-order optimizer that uses a light-weight estimate of the diagonal Hessian as the pre-conditioner. The update is the moving average of the gradients divided by the moving average of the estimated Hessian, followed by element-wise clipping. The clipping controls the worst-case update size and tames the negative impact of non-convexity and rapid change of Hessian along the trajectory. Sophia only estimates the diagonal Hessian every handful of iterations, which has negligible average per-step time and memory overhead. On language modeling with GPT-2 models of sizes ranging from 125M to 770M, Sophia achieves a 2x speed-up compared with Adam in the number of steps, total compute, and wall-clock time. Theoretically, we show that Sophia adapts to the curvature in different components of the parameters, which can be highly heterogeneous for language modeling tasks. Our run-time bound does not depend on the condition number of the loss.

top 1 comments

sorted by: hot top controversial new old

[–] nsa@kbin.social 1 points 1 year ago

Research into efficient optimization techniques seems pretty important given the scale of LLMs these days. Nice to see a second-order approach that achieves reasonable wall-clock improvements.