Agree. What area's in AI you think would be good bets in the next 10 years in Industry or Academia?
If you're a grad student, it's hard to make a mark in deep learning because it's so overcrowded. You're better off working in another area of AI.
-
-
-
Other subfields of ML, analogical reasoning, multiagent systems, etc.
- Näytä vastaukset
Uusi keskustelu -
-
-
IMO grad students need to be trained to be useful, not necessarily do something new for the sake of publication count. Academic incentive structure needs improvement.
Kiitos. Käytämme tätä aikajanasi parantamiseen. KumoaKumoa
-
-
-
Other subfields of ML, analogical reasoning, multiagent systems, etc. would be better
Kiitos. Käytämme tätä aikajanasi parantamiseen. KumoaKumoa
-
-
-
Deep learning is awesome but its taken focus away from research into other approaches to machine intelligence ...
Kiitos. Käytämme tätä aikajanasi parantamiseen. KumoaKumoa
-
-
-
if you're going to work on deep learning, try to research 'position encodings' in transformers. it is rife with adhoc solutions and needs to be brought into a general framework across modalities
Kiitos. Käytämme tätä aikajanasi parantamiseen. KumoaKumoa
-
-
-
Why not learn ML/DL and work on application in ones area of liking? After all, a sound footing in basics (via course work) and then applied research will bring much needed work force in these areas. Besides, anyone can make it good (basics + brain power) even if it's ONLY DL.
Kiitos. Käytämme tätä aikajanasi parantamiseen. KumoaKumoa
-
-
-
If I followed this kind of advice I'd be miserable now. There is a severe shortage of expertise and far too many pretenders/influencers. If you are in deep learning and become competent you will go far as you will be better than most people calling themselves experts.
-
Now this…sounds like amazing advice Preciate the words because he tweaking
Keskustelun loppu
Uusi keskustelu -
-
-
That's assuming you want to get a reputation in academia If instead you want to do useful things for people, learn about useful tools and use them, be flexible
Kiitos. Käytämme tätä aikajanasi parantamiseen. KumoaKumoa
-
Lataaminen näyttää kestävän hetken.
Twitter saattaa olla ruuhkautunut tai ongelma on muuten hetkellinen. Yritä uudelleen tai käy Twitterin tilasivulla saadaksesi lisätietoja.