Update Teaser - Gametes

Here anything Cell Lab related that doesn't fit into the other topics can be discussed.
User avatar
bwisialo
Posts: 2955
Joined: Sat Aug 15, 2015 10:22 pm

Re: Update Teaser - Gametes

Postby bwisialo » Tue May 30, 2017 11:19 pm

@Nayus -- Do you mean a setting to control the amount of change a mutation causes, or do you mean the rate at which mutations occur?

Both options would allow you to turn mutations off at 0. A setting for mutation rate would be interesting. You could also have this setting subject to mutation under radiation, as Petter mentioned. It seems like the amount of change is of a different nature depending on which setting you are looking at: you wouldn't want to apply a percentage across the board.
amor fati

User avatar
Nayus
Posts: 570
Joined: Thu Oct 06, 2016 2:03 am
Location: Argentina

Re: Update Teaser - Gametes

Postby Nayus » Wed May 31, 2017 12:07 am

The value would be the amont of error the gamene is allow to produce, as a maximum in either side.

I'll call this setting "Variation Probability" of "Limit of Variation Probability" for now.

0 of Variation Probability would produce no changes, ever. All the offspring will be exactly the same as the modes of their parent modes.

and you would go up. For example a 5% of that value will let some settings change up to a 5% different. Let's say you can have certain setting with a minimum value of 0 and a maximum of 30. If the parent Mode has, let's say 15 in a moment and the gamete is set to a 5% Variation Probability Limit. 5% of 30 (changing from 0 to 30 would be a 100%) is 1.5. Meaning that the child can vary +-1.5 value from before; so about 13.5 as minimum change and 16.5 as maximum. It's a very small change because 5% is a small value.

If it were 50%, then you could offspring changing from 15 to 30 in one generation.

The important thing is that is a probability. I don't know how often the changes should be in place.. But the point is that sometimes the settings will change very little or nothing at all even though the value is greater than 0%

User avatar
Nayus
Posts: 570
Joined: Thu Oct 06, 2016 2:03 am
Location: Argentina

Re: Update Teaser - Gametes

Postby Nayus » Wed May 31, 2017 12:09 am

The "how often" thing is now a problem.....

I'm thinking that it could be a second value, but it could be too much right? I don't know if the gametes have exclusive settings on their own.

User avatar
bwisialo
Posts: 2955
Joined: Sat Aug 15, 2015 10:22 pm

Re: Update Teaser - Gametes

Postby bwisialo » Wed May 31, 2017 12:32 am

Right. I don't think you'd ever want one percentage value applied to all settings. That's why I broke down the range of mutation differently for different settings. If you have a 10% change applied to Nutrient priority in order to have changes of less than or equal to +/- 1, that doesn't mean you want Neurocyte channel values for "a" and "b" to change by less than or equal to +/-8 (or +/- 16; the range is -80 to 80). Oscillation presets have values of "a" less than 10 and b=0 in the first two channels. I also don't think percentages would be the way to go with angles.

PS edit: Mutation rate 0 would likewise mean no changes ever.

And, yes, values would always be able to change a small value or large value randomly within the allowed mutation range.
amor fati

User avatar
Nayus
Posts: 570
Joined: Thu Oct 06, 2016 2:03 am
Location: Argentina

Re: Update Teaser - Gametes

Postby Nayus » Wed May 31, 2017 12:53 am

Yes I agree but if we started to pick which values are more 'vulnerable' to changes, to find a word; or are affected more by the changes, wouldn't that betray the sense of reproduction? It would be very specifically picked to work in order to certain directions we want them to work, mainly the Evolution from Dumb to smart.
If we make certain values 'harder' to get changed, it wouldn't feel like true evolution anymore, I believe.

User avatar
bwisialo
Posts: 2955
Joined: Sat Aug 15, 2015 10:22 pm

Re: Update Teaser - Gametes

Postby bwisialo » Wed May 31, 2017 1:13 am

No, I don't mean you'd get to say that certain settings can mutate and others not. I just mean that the range of change for Neurocyte "a" and "b" should be of a smaller factor than the range for Nutrient priority. I tried do lay out how different settings could be done differently. Not that you'd choose some settings and not others.
amor fati

User avatar
bwisialo
Posts: 2955
Joined: Sat Aug 15, 2015 10:22 pm

Re: Update Teaser - Gametes

Postby bwisialo » Wed May 31, 2017 1:16 am

I think this is what Petter meant when he wrote:

It's hard to judge how to do this though. Should we just allow +-5% difference, or maybe +-50%? I think that will be very important and it will depend on the property.
amor fati

User avatar
bwisialo
Posts: 2955
Joined: Sat Aug 15, 2015 10:22 pm

Re: Update Teaser - Gametes

Postby bwisialo » Wed May 31, 2017 1:22 am

Oops, I slightly misunderstood your objection. No, I don't think different sorts of factors for different settings is contrary to reproduction or evolution. Eg,the values for different settings have different sorts of scales. Eg, some are more linear and others are more exponential in their effects.

Edit: It's not that different factors would be designed for certain outcomes. It's that there would be a certain sense of proportionality in how different settings mutate: 5% of one setting is not necessarily functionally equivalent to 5% in another.

Sorry for all the edits.
amor fati

User avatar
Nayus
Posts: 570
Joined: Thu Oct 06, 2016 2:03 am
Location: Argentina

Re: Update Teaser - Gametes

Postby Nayus » Wed May 31, 2017 1:26 am

Exactly, now I get why percentages wouldn't fix it either.
I believed the opposite, that a percentage would put things in equal footing, but that makes sense. 80 in the Neurocyte is a bit much, while the lower values act as sort of logarithmic-ish.

But I still think it should be avaliable to choose if the changes are possible or not. Even if it's something very limited as a Check Box that states "Allow Copying Errors". It would be nice to set how much it's allowed, but still.

Because both separate options, of having AND not having errors seems restrictive and it will produce problems.

User avatar
bwisialo
Posts: 2955
Joined: Sat Aug 15, 2015 10:22 pm

Re: Update Teaser - Gametes

Postby bwisialo » Wed May 31, 2017 1:33 am

Sure. I agree that you should be able to turn the mutations on/off. Supposing we get these mutations at all. :) But a setting for mutation rate could be better, sort of like radiation on the Substrate tab. Off is 0, and you can decide how frequently / fast the genome is changing.
amor fati

Return to “General Discussion”

Who is online

Users browsing this forum: No registered users and 1 guest