Summary of Random Scaling and Momentum For Non-smooth Non-convex Optimization, by Qinzi Zhang et al.
Random Scaling and Momentum for Non-smooth Non-convex Optimization
by Qinzi Zhang, Ashok Cutkosky
First submitted to arxiv on: 16 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Optimization and Control (math.OC)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed algorithm combines stochastic gradient descent with momentum (SGDM) and an exponentially distributed random scalar to optimize highly irregular loss functions. This modification closes the gap between classical SGDM analysis and its application in practice, providing optimal convergence guarantees. The approach is derived from a general framework for converting online convex optimization algorithms to non-convex optimization algorithms. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A team of researchers found a way to make training neural networks more efficient. They modified an existing algorithm called stochastic gradient descent with momentum (SGDM) by adding a random number at each step. This simple change makes the algorithm work well even when the loss function is not convex or smooth, which is often the case in practice. The new approach has better guarantees for how quickly it will converge to a solution. |
Keywords
» Artificial intelligence » Loss function » Optimization » Stochastic gradient descent
