People who think evolution works through random mutations and selection need to explain how intelligent life appeared using nothing else. Clearly, any optimization process is more efficient if it uses some sort of gradient estimation.https://twitter.com/pmddomingos/status/1463004789269151744 …
The first is true, but they have at least a sketch of it. The second is false (no free lunch for optimization). Neither lets you off the hook. Gradient descent is not enough for scientific discovery, and therefore also not for human-level AI.
-
-
Whatever pulls the future into the present by the way of emergence with the experience of the past is how intelligence develops imo.
Kiitos. Käytämme tätä aikajanasi parantamiseen. KumoaKumoa
-
-
-
If intelligence is helpful in survival and propagation then evolution will select for it. Gradient decent toward intelligence leaves you stuck at a local minimum.
-
Evolution, as a hill climbing algorithm, also only ever leaves you stuck on local extrema.
- Näytä vastaukset
Uusi keskustelu -
-
-
Currently doing my Ph.D with the goal of convincing people that it's atleast sufficient for scientific discovery.
-
Don't confuse curve-fitting with scientific discovery.
- Näytä vastaukset
Uusi keskustelu -
-
-
Is genetic programming enough for scientific discovery?
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.