Evolutionary Algorithms and ID

[Giltil (Gilbert Thill)]

@Jordan

My question is, is this “500 bits” criteria (which at this point seems completely arbitrary to me, but that’s probably for another post) for intelligent design a cumulative thing or for each individual step? So if @Mercer is calculating 26 bits of FI in a single recombination event, will 20 recombination events “make it” or would we have to find a single “thing” that has 500 bits of FI?

I’ve already answered you at 29 but I thought you might be interested by @gpuccio’s insightful analysis of this issue below:

Gpuccio: Functional Information Methodology

OK, let’s clarify this. 10 objects with 50 bits of FI each are not, in any way, 500 bits of FI. 10 objects with 50 bits of FI each are 500 bits of FI only if those 10 exact objects are needed to give some new defined function. Let’s see the difference. Let’s say that there is a number of possible functions in a genome that have, each of them, 50 bits of FI. Let’s call the acquisition of the necessary information to get one of those functions “a success”. These functions are the small safes in my example. The probability of getting a success in one attempt is, of course, 1:2^50. How many attempts are necessary to get at least one success? This can be computed using the binomial distribution. The result is that with 2^49 attempts we have a more than decent probability (0.3934693) of getting at least one success. How many attempts are necessary to have a decent probability of gettin at least 10 successes, each of them with that probability of success, each of them with 50 bits …

[glipsnort (Steve Schaffner)]

@Giltil

I’ve already answered you at 29 but I thought you might be interested by @gpuccio’s insightful analysis of this issue below:

Gpuccio: Functional Information Methodology Scholars

What is the search space for 100 1-bit safes? What is the search space for on 100-bit safe? Ditto for the target spaces.

[Chris_Falter]

I think @Roy knows what he’s talking about. In the closely related field of machine learning, we have reinforcement learning with generative neural networks. I’ll call it RLWGNN for short. They maximize a reward function, but the details of strategy and tactics are unspecified; they are what the RLWGNN.does.

[colewd (Bill Cole)]

@glipsnort

What is the search space for 100 1-bit safes? What is the search space for on 100-bit safe? Ditto for the target spaces.

Functional information is estimated for a single specific function. In the first case you have 100 independent functions.

[glipsnort (Steve Schaffner)]

@colewd

Functional information is estimated for a single specific function. In the first case you have 100 independent functions.

What are the target and search spaces? How many sequences are in each?

[colewd (Bill Cole)]

@glipsnort

What are the target and search spaces? How many sequences are in each?

The search space for a 100 bit safe is 100 bits assuming one solution. The search space for a 1 bit safe is 1 bit assuming two possibilities only.

For there to be a 100bit search space in the case of 100 1bit safes all 100 safes need to have the proper code for the safes to open. In this case all 100 safes point to a single function of 100 1bit safes opening.

[glipsnort (Steve Schaffner)]

@colewd

The search space for a 100 bit safe is 100 bits assuming one solution. The search space for a 1 bit safe is 1 bit assuming two possibilities only.

For there to be a 100bit search space in the case of 100 1bit safes all 100 safes need to have the proper code for the safes to open. In this case all 100 safes point to a single function of 100 1bit safes opening.

What are the target and search spaces for 100 1-bit safes? How many sequences are in each?

[colewd (Bill Cole)]

@glipsnort

What are the target and search spaces for 100 1-bit safes? How many sequences are in each?

There are 100 targets not one.

[glipsnort (Steve Schaffner)]

@colewd

There are 100 targets not one.

So? There’s still a target space and a search space. What are they? How many sequences are in each?

[Giltil (Gilbert Thill)]

@Jordan

My question is, is this “500 bits” criteria (which at this point seems completely arbitrary to me, but that’s probably for another post

This «500 bits » criteria is not arbitrary. 500 bits corresponds to the information content of an event having a probability of 1 out of 10^150 of occurring. This is what Dembski called the universal probability bound, which he defines as follow:
« A degree of improbability below which a specified event of that probability cannot reasonably be attributed to chance regardless of whatever probabilitistic resources from the known universe are factored in ».
So, according to ID theory, a single event of that probability (1/10^150), or put differently, an single event producing 500 bits of FI cannot be attributed to chance. In that case, design is the right explanation.

[colewd (Bill Cole)]

@glipsnort

So? There’s still a target space and a search space. What are they? How many sequences are in each?

Not unless there is one defined function. Until that is defined you cannot answer your question relative to functional information. Are there 100 independent searches going on or 1 search where 100 safes are involved?

[Mercer (John Mercer)]

@BWhite

How did anyone determine that the immune system evolved by blind, mindless and purposeless processes?

We’re not talking about how it evolved. We’re talking about how it creates new information in real time using variation and selection.

[Mercer (John Mercer)]

@Giltil

The answer is that 20 events will NOT « make it » and we would have to find a single « thing » that has 500 bits of FI.

Why? Because you say so? Aren’t those goalposts getting heavy?

Next, you’ll be equivocating on the meaning of “thing.”

@Giltil

So, according to ID theory,

There is no “ID theory,” Gil. A theory is a hypothesis with a long track record of successful predictions. You’ve merely got a hypothesis that you’re clearly unwilling to test rigorously.

[Mercer (John Mercer)]

@colewd

FI is function specific. If the single “thing” is a function then you are correct.

I don’t see how “function” is any less arbitrary than “thing.”

Gotta make sure those goalposts are very mobile, right?

[glipsnort (Steve Schaffner)]

@colewd

Not unless there is one defined function. Until that is defined you cannot answer your question relative to functional information. Are there 100 independent searches going on or 1 search where 100 safes are involved?

The claim is that FI is lower for the 100 safe case. That must mean that you can calculate a value for FI. Just tell me what it is, for whatever search you think is appropriate, and how it’s calculated.

ETA: The target is clear: open all of the safes present, whether 1 or 100.

[Mercer (John Mercer)]

@Jordan

My question is, is this “500 bits” criteria (which at this point seems completely arbitrary to me, but that’s probably for another post) for intelligent design a cumulative thing or for each individual step? So if @Mercer is calculating 26 bits of FI in a single recombination event, will 20 recombination events “make it” or would we have to find a single “thing” that has 500 bits of FI?

Then we can easily set the “thing” or “function” as the humoral immune response, because based on a huge amount of data, there are easily 100 clones doing this function together, in parallel.

There’s also the cellular immune response, which does a similar, yet different thing.

Come on, Jordan, you’re a scientist. These guys have absolutely no intention of testing a hypothesis. @gpuccio even told us that his goal was polemic.

[Dan_Eastwood]

FYI: There is extended discussion of the Safes and Thieves example here:

Comments on Gpuccio: Functional Information Methodology

@gpuccio here is a reference point for the discussion. This definition is from Hazen and Szostak. I(Ex) = ?log2[F(Ex)], where F(Ex) is the fraction of all possible configurations of the system that possess a degree of function ? Ex. Functional information, which we illustrate with letter sequences, artificial life, and biopolymers, thus represents the probability that an arbitrary configuration of a system will achieve a specific function to a specified degree. In each case we observe evidence for several distinct solutions with different maximum degrees of function, features that lead to steps in plots of information versus degree of function.

And here:

Information is Additive but Evolutionary Wait Time is Not

Actually the opposite. he thinks it must happen all at once. He needs that, because it MUST happen at all at once for his wait time = 2^FI claim to be correct. Decomposable and parallel are distinct concepts. A decomposable, but non-parallel, problem would change this: To this: In fully decomposible system (100 safes) the wait time is about 100 * 2 . Here one thief is working, solving each safe one at a time. Exactly, so this last case is just not relevant. There is at least two other simplifications in this case study, which if we fixed, it would even further decrease wait time to be sublinear.

[Rumraket (Mikkel R.)]

@Giltil

This «500 bits » criteria is not arbitrary. 500 bits corresponds to the information content of an event having a probability of 1 out of 10^150 of occurring. This is what Dembski called the universal probability bound, which he defines as follow:
« A degree of improbability below which a specified event of that probability cannot reasonably be attributed to chance regardless of whatever probabilitistic resources from the known universe are factored in ».
So, according to ID theory, a single event of that probability (1/10^150), or put differently, an single event producing 500 bits of FI cannot be attributed to chance. In that case, design is the right explanation.

But as we’ve been over many times, this assumes the thing in question was produced in a single event, as opposed to an accumulation of smaller events, some of which receive feedback from selection.

[Rumraket (Mikkel R.)]

@BWhite

Blind, mindless and purposeless processes do not search.

Then all arguments about how difficult it would be to “find” functions with blind searches are irrelevant to evolution and therefore fail as arguments against it. You should contact Dembski and tell him you’ve found a fatal flaw in his ideas.

[Giltil (Gilbert Thill)]

@Mercer

How did anyone determine that the immune system evolved by blind, mindless and purposeless processes?

We’re not talking about how it evolved. We’re talking about how it creates new information in real time using variation and selection.

As shown below by @gpuccio, the new information you are talking about, although random in some aspects, is in fact produced by « functional protein engineering embedded in a complex biological system ». IOW, design is involved at all stages in the production of new information by the IS.
https://uncommondescent.com/intelligent-design/antibody-affinity-maturation-as-an-engineering-process-and-other-things/