# Adaptive learning rates and parallelization for stochastic, sparse, non-smooth gradients

arXiv: https://arxiv.org/abs/1301.3764 · HTML: https://mlsanity.com/paper/1301.3764/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Jan 2013.

## Built on by (most cited first, 1 of 1)

- [Adaptive Sampling Quasi-Newton Methods for Zeroth-Order Stochastic Optimization](https://mlsanity.com/paper/2109.12213/) (2021)

## Abstract (arXiv)

Recent work has established an empirically successful framework for adapting learning rates for stochastic gradient descent (SGD). This effectively removes all needs for tuning, while automatically reducing learning rates over time on stationary problems, and permitting learning rates to grow appropriately in non-stationary tasks. Here, we extend the idea in three directions, addressing proper minibatch parallelization, including reweighted updates for sparse or orthogonal gradients, improving robustness on non-smooth loss functions, in the process replacing the diagonal Hessian estimation procedure that may not always be available by a robust finite-difference approximation. The final algorithm integrates all these components, has linear complexity and is hyper-parameter free.