4 ms·
Basically everything you interact with on a daily basis involves (deep) machine learning. Everything from advertising to logistics to drug discovery to credit e
by maaku 11y ago
Basically everything you interact with on a daily basis involves (deep) machine learning. Everything from advertising to logistics to drug discovery to credit evaluations to packet routing on the internet either involves deep learning in its operation, or is presently implemented with some optimal strategy that was discovered via deep learning. Machine learning runs the world.
As for the business opportunity, if you are in one of these industries then one of the prime ways you differentiate from your competitors is how well your algorithm works. And how well it works depends a great deal on how much training data you are able to feed into it, which in this era of information saturation is basically a hardware limitation.
Give me a 10x more powerful machine learning system, and I'll give you a few basis points advantage over the competition, and that'll make you, not them, the dominant player.
- hueving 11y agoPacket routing? If there is an example of this, it is very far from the norm. Routing on the Internet is driven first and foremost by peering arrangements.
- maaku 11y agoNow ask yourself how decisions are made regarding new infrastructure deployments. But what I actually had in mind was QoS.
- mryan 11y agoWho is using deep learning for packet routing and QoS decisions?
- pilooch 11y agoNo one I know of, from the last time I did check on this topic. However, it seems possible to plug ML on top of something like OpenFlow. PM me if interested to look this up further.
- mryan 11y agoThanks. I was mostly interesting in seeing a source for maaku's statement of "Now ask yourself how decisions are made regarding new infrastructure deployments", as it seemed pretty condescending to me.
- hueving 11y agoNew infrastructure deployments is not "packet routing" and QoS is usually as strongly tied to policy as BGP is. If you have an example of QoS relying on ML, I would be interested in reading more.
- cscurmudgeon 11y agoEven in AI, deep learning is not the state of the art in the majority of domains. E.g., in QA Watson is still the best in Jeopardy style general QA. In general game play, statistical methods rule, etc.