# Exploring loss function topology with cyclical learning rates

arXiv: https://arxiv.org/abs/1702.04283 · HTML: https://mlsanity.com/paper/1702.04283/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Feb 2017.

## Built on by (most cited first, 11 of 11)

- [Predicting Grokking Long Before it Happens: A look into the loss landscape of models which grok](https://mlsanity.com/paper/2306.13253/) (2023)
- [Stochastic Weight Averaging Revisited](https://mlsanity.com/paper/2201.00519/) (2022)
- [Sparse is Enough in Fine-tuning Pre-trained Large Language Models](https://mlsanity.com/paper/2312.11875/) (2023)
- [Unveiling the Basin-Like Loss Landscape in Large Language Models](https://mlsanity.com/paper/2505.17646/) (2025)
- [Data efficiency and extrapolation trends in neural network interatomic potentials](https://mlsanity.com/paper/2302.05823/) (2023)
- [Fast-Convergent Federated Learning via Cyclic Aggregation](https://mlsanity.com/paper/2210.16520/) (2022)
- [Bringing the Discussion of Minima Sharpness to the Audio Domain: a Filter-Normalised Evaluation for Acoustic Scene Classification](https://mlsanity.com/paper/2309.16369/) (2023)
- [Regime-Switching Langevin Monte Carlo Algorithms](https://mlsanity.com/paper/2509.00941/) (2025)
- [Cyclic and Randomized Stepsizes Invoke Heavier Tails in SGD than Constant Stepsize](https://mlsanity.com/paper/2302.05516/) (2023)
- [LossLens: Diagnostics for Machine Learning through Loss Landscape Visual Analytics](https://mlsanity.com/paper/2412.13321/) (2024)
- [Visualization and Analysis of the Loss Landscape in Graph Neural Networks](https://mlsanity.com/paper/2509.11792/) (2025)

## Abstract (arXiv)

We present observations and discussion of previously unreported phenomena discovered while training residual networks. The goal of this work is to better understand the nature of neural networks through the examination of these new empirical results. These behaviors were identified through the application of Cyclical Learning Rates (CLR) and linear network interpolation. Among these behaviors are counterintuitive increases and decreases in training loss and instances of rapid training. For example, we demonstrate how CLR can produce greater testing accuracy than traditional training despite using large learning rates. Files to replicate these results are available at https://github.com/lnsmith54/exploring-loss