From f42ffae7f571b826ab7fad0eb77ca78f7ffa9d31 Mon Sep 17 00:00:00 2001 From: Jim Martens Date: Wed, 23 May 2018 15:43:46 +0200 Subject: [PATCH] [NN] Added short descriptions of the work of Shmelkov, Velez and Kirkpatrick Signed-off-by: Jim Martens --- neural-networks/seminarpaper.tex | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/neural-networks/seminarpaper.tex b/neural-networks/seminarpaper.tex index 685e42d..40755dc 100644 --- a/neural-networks/seminarpaper.tex +++ b/neural-networks/seminarpaper.tex @@ -294,10 +294,11 @@ would then serve as teacher for the second network which models the neocortex and is responsible for generalizing. In the time between 1999 and 2018 more work was done with regards to catastrophic -forgetting. Most recently the work of Kirkpatrick\cite{Kirkpatrick2017}, -Velez\cite{Velez2017} and Shmelkov\cite{Shmelkov2017} has to be named. The main -focus here was a brief introduction of catastrophic forgetting and therefore -these newest approaches will not be described here. +forgetting. Most recently the work of Kirkpatrick\cite{Kirkpatrick2017}, who +slows down the learning for weights of older tasks, +Velez\cite{Velez2017}, whose work will be showcased later, and +Shmelkov\cite{Shmelkov2017}, who introduces a loss function that intends to keep +catastrophic forgetting at bay, has to be named. \section{Plasticity} \label{sec:plasticity}