Deep learning is just another way of doing statistics.
-it makes strong assumption
-you have large amounts of data
-classifying into categories.
-test data aren't far from training examples
@GaryMarcus #ttiintelligencepic.twitter.com/7xcFRNE3NI
This is the legacy version of twitter.com. We will be shutting it down on June 1, 2020. Please switch to a supported browser, or disable the extension which masks your browser. You can see a list of supported browsers in our Help Center.
FOLLOWS YOU. Artificial Intelligence, Cognitive Architectures, Computation. The goal is integrity, not conformity.
You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
Add this Tweet to your website by copying the code below. Learn more
Add this video to your website by copying the code below. Learn more
By embedding Twitter content in your website or app, you are agreeing to the Twitter Developer Agreement and Developer Policy.
| Country | Code | For customers of |
|---|---|---|
| United States | 40404 | (any) |
| Canada | 21212 | (any) |
| United Kingdom | 86444 | Vodafone, Orange, 3, O2 |
| Brazil | 40404 | Nextel, TIM |
| Haiti | 40404 | Digicel, Voila |
| Ireland | 51210 | Vodafone, O2 |
| India | 53000 | Bharti Airtel, Videocon, Reliance |
| Indonesia | 89887 | AXIS, 3, Telkomsel, Indosat, XL Axiata |
| Italy | 4880804 | Wind |
| 3424486444 | Vodafone | |
| » See SMS short codes for other countries | ||
This timeline is where you’ll spend most of your time, getting instant updates about what matters to you.
Hover over the profile pic and click the Following button to unfollow any account.
When you see a Tweet you love, tap the heart — it lets the person who wrote it know you shared the love.
The fastest way to share someone else’s Tweet with your followers is with a Retweet. Tap the icon to send it instantly.
Add your thoughts about any Tweet with a Reply. Find a topic you’re passionate about, and jump right in.
Get instant insight into what people are talking about now.
Follow more accounts to get instant updates about topics you care about.
See the latest conversations about any topic instantly.
Catch up instantly on the best stories happening as they unfold.
Deep learning is just another way of doing statistics.
-it makes strong assumption
-you have large amounts of data
-classifying into categories.
-test data aren't far from training examples
@GaryMarcus #ttiintelligencepic.twitter.com/7xcFRNE3NI
Cognition is also just another way of doing statistics
doing statistics is just one (important) aspect of cognition. at other extreme, would you call formal logic statistics?
Ha! Statistics has absolutely nothing to do with cognition. As Judea Pearl once correctly said, people are not probability thinkers but cause-effect thinkers. The brain uses a winner-take-all method to handle uncertainty.https://medium.com/@RebelScience/the-brain-is-not-probabilistic-82ac88e0a3bf …
I disagree here. I think the brain uses probabilistic assumptions. I believe nested sampling would work best as the Monte Carlo Markov Chain Method, which in cognitive psychology is the assumed method of the brain.
Well, I will not argue my point further other than to say this: none of us see 1%, 15% or 90% cow. It's either cow or no cow.
If you imagine or perceive a cow, you will have to represent a superpositional state that approximates probabilities. But when you access that representation for reasoning or prediction, you will get a collapsed version (because a probability cloud shaped cow is very unlikely).
It's impossible to represent every possible cow in memory seen from different angles because the number is infinite. We only maintain high level bits of what a cow looks like. And yet we can recognize a cow image mixed with > 90% noise. And when we do, we are 100% certain.
We can represent the angle as indeterminate. If you need to reason about the cow, you generate it with a definite angle but put low confidence on it, so the angle is not clamped down and can be regenerated if the constraints change.
Other representations use multiple alternative prototypes to cover the feature space. What fools you is that attentional access uses lateral inhibition to collapse it into one.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.