Real-world applications must balance machine learning with rule-based extraction pipelines
Antoine Bordes · Artificial Intelligence at Facebook // Antoine Bordes, Facebook [FirstMark's Data Driven] · Nov 9, 2016 · at 24:45
Antoine Bordes, AI Research Scientist at Facebook AI Research, explains the practical engineering trade-off between pure machine learning and traditional rule-based pipelines for real production software.
0:00 / 0:31exact quote · 31.5s
▶ Watch the full episode on YouTube → 720p mp4 · rendered on demand · StarZero watermark
“In terms of developing real application, I think you should really try to balance both, because it's true that in the example I showed, If you have a very good process to basically already isolate all the entities and if you want basically to use something like this in your product, yeah, you don't want to force yourself to have a network that has to rediscover something that you can actually extract by yourself using your pipeline. That would be completely another kill. So so yeah, I think when you basically want to have something that works well, you basically have to mix both.”
quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →
More from Antoine Bordes
Bordes: 2016 image recognition AI uses Yann LeCun's 1993 neural architecture
“This is exactly the same architecture of what's being used right now.”
Bordes: Direct machine reading will eventually match IBM Watson's performance
“The goal is that, ah, eventually, by trying to understand the text directly, you can actually at least equal, all the information by Watson is in text, in free text.”
Most 2016 production computer vision systems lack bounding box capabilities
“This is, right now, I would say that the best systems are doing this in production. Most of them are doing this. And some are not even doing the boxes.”
Bordes predicts large AI models will run on mobile devices by 2017
“And next year we will run them on mobile, so.”