Systematic analysis of many recent neural #recsys shows lack of reproducibility and poor performance even when compared to simple heuristics 😨 https://t.co/v2iPua3UMM
— Xavier 🎗️ (@xamat) July 17, 2019
Systematic analysis of many recent neural #recsys shows lack of reproducibility and poor performance even when compared to simple heuristics 😨 https://t.co/v2iPua3UMM
— Xavier 🎗️ (@xamat) July 17, 2019
Relevant to this blog post I wrote a while ago: https://t.co/lyw14bLMZc
— Rachael Tatman (@rctatman) July 29, 2019
Unfortunately it looks like even more evidence that common methods of ML algorithm evaluation have some pretty big flaws. 😢 https://t.co/qgQ3DTY6FF
"On the Difficulty of Evaluating Baselines: A Study on Recommender Systems" [PAPER] #recsys https://t.co/4NqSfpP7al
— Xavier 🎗️ (@xamat) July 30, 2019
"With a [...] vanilla MF [...], we are not only able to improve upon the reported results for this baseline but even outperform [...] any newly proposed method" 😱