From 7e2154830055b30189a928cbffa5859c1be0cf5e Mon Sep 17 00:00:00 2001 From: Jeffrey Ede Date: Tue, 25 Feb 2020 14:35:52 +0000 Subject: [PATCH] Update README.md --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 5fc9b1b..d901527 100644 --- a/README.md +++ b/README.md @@ -8,7 +8,7 @@ This repository contains source code for CIFAR-10 supersampling experiments with

-Example learning curves showing that ALRC stabilizes learning by preventing loss spikes at unstably high learning rates, and otherwise has little effect. Learning curves are 500 iteration boxcar averaged. Results hold for low and high-order loss functions, different batch sizes and different optimizers. +Example learning curves for stable and unstably high learning rates. ALRC stabilizes learning by preventing loss spikes and otherwise has little effect. Learning curves are 500 iteration boxcar averaged. Results hold for low and high-order loss functions, different batch sizes and different optimizers. # Description