On Mon, Oct 19, 2020, 11:05 PM <[email protected]> wrote:

> On Monday, October 19, 2020, at 10:44 PM, Matt Mahoney wrote:
>
> Shannon information assumes the probability distribution is given and
> tells you how to code it. Algorithmic information tells you the probability
> distribution, and that you can't compute it.
>
>
> I don't understand, if 'dog barked' is more common than 'dog slept' and
> prompted with 'dog __' I predict barked more than slept, where does what
> you're saying fit into text prediction? You either predict or you don't....
>

Shannon says you can code "barked" in -log2(P(barked|dog)) bits.
Algorithmic information tells you how to compute P.


------------------------------------------
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T6761a13445e5864b-Mda198db3bc6a7177d9b179c5
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Reply via email to