Still cranking away on my neural network project

January 20th, 2018

Here’s a screen capture to tease you (and give you the sense that I’m doing something interesting over here)

I always test my code…

January 8th, 2018

On my office wall, I have a poster that has a picture of me with “I always test my code.. in production. When I’m coding myself, there’s nowhere else to test.”

This is about to not be true. I am building a genuine, very real, not in any way a delusion neural network debugger.

I will be posting a link to the git repo once it’s approximately functional so you can all appriciate the cleverness that goes into this. (Or laugh at me ;))

In the meantime, I leave you with this image of that poster.

Luck vs Choice?

December 4th, 2017

So, one of the questions I tend to ask myself, as I talk to people who can’t troubleshoot simple machinery, is to what extent did I get lucky and to what extent have my choices led me to where I am?

It’s a worthwhile question. Did a simple throw of the genetic dice, or the path that I was led down, lead to me being capable of understanding almost any human-made system? Or is it my repeated choices to read, to study, to attempt to fix things even when I don’t actually know how, to ask questions of other people, to – not to put too fine a point on it – continuously learn and evolve over the course of my life?

Sometimes I get incredibly frustrated when talking to people who are not as capable as I am and who repeatedly insist that they can’t do something. Pretty much everything built by humans can be understood by humans and fixed by humans. And I wonder, is this a choice they’re making? Do people choose to be less capable than they are biologically able to be? Sometimes it feels extremely choice-driven – and yet, I am not at all clear whether it is or not. Re: previous discussions on free will, I think that not everyone has as large a list of options in their ‘what can I choose to do right this second’ list as I do, and I think some of that is that the more you learn, the larger your free will window becomes. So people who haven’t been imbued with a can-do attitude and experienced validation of that attitude literally can’t choose to believe that they can i.e. troubleshoot their car.

I have also seen people create large numbers of imaginary obstacles for themselves before they ever even attempt the job at hand. Now, I should mention that I think memetic disempowerment is a systematic problem with humanity – recently someone reminded me of the quote “All have sinned and fallen short of the glory of God” which I think is a *excellent* example of memetic disempowerment and one of the many reasons that Christianity deserves relegated to the dustbin of history. Yes, sure, believe you’re going to fail before try! That’ll help! I also think that there is a fair amount of memetic disempowerment that goes on in our educational system – repeatedly grading people is not likely to help them feel empowered unless they happened to start out at the top of the ladder – and in our consumer-driven world, since after all if you feel empowered enough you might not buy $WHATEVER.

I am sure I also create imaginary obstacles for myself, and I’m sure that I have also frustrated many people in the past in ways similar to how I am sometimes frustrated by others now. I do wonder, though, how much of this is a choice and how much of it is directed by the wiring and memetic programming?

Another question is, what do we owe those who can’t? The political powers that be would, it would seem, like to throw anyone who isn’t extremely capable in all areas under the bus – and I assume that sooner or later this will include me, since if we keep raising the hurdle, sooner or later I will not be able to jump it. It’s clear that if we wanted to feed and house everyone we could, but also that we feel warm and fuzzy about patting ourselves on the back as we throw those who are less capable under the bus. Personally, I think we should try and feed and clothe and house everyone – in fact, give everyone everything they want, to the extent of our capability – although there are those who argue that we wouldn’t enjoy things if we didn’t have to struggle for them.

I don’t know. Rereading this post, I feel kind of like it paints me as a awful person, and that isn’t really my intention at all.

From a facebook discussion : free will

November 23rd, 2017

Well, the problem I have with saying I have free will is multifold. A: I am not sure I exist. “I” as a single entity might well be a illusion since I appear to be a cooperating collection of subnets, and experiments like cutting the corpus callosum argue strongly that I am not a single ego, that this is a illusion. B: I am not sure, if I do exist, that I’m not deterministic. Experimenting with artificial neural networks, I note that they tend strongly towards the deterministic unless measures are taken to keep them from being deterministic. C: I am not sure, if I do exist and am not deterministic, that it is free agency and not a RNG or random noise that is guiding my actions. And yet, the idea that I am a person wandering around taking actions of my own free will is very compelling. Especially when I start discussing the matter which seems very meta

Trippy Hippie Meditation Music

November 15th, 2017

More movie/atmospheric stuff

Thought

SimSheer

November 14th, 2017

So, one of the things I’ve been learning about is ANNs. I’ve tried playing with several different frameworks and several different topologies, and one of the ones I’ve been playing with is Darknet.

I’ve been trying to train a Darknet RNN on a corpus generated from all the text in my blog. So far the results have been less than stellar – I think I need a bigger neural network than I’ve been using, and I think in order to do that I need a bigger GPU because I’m running out of patience. I was astonished to discover >1 teraflop GPUs are now in my price range, so I’ve ordered one.

I’m hoping soon to have simSheer available as a php endpoint that people can play with. All of this is building up to using Darknet for some other purposes, such as image recognition.

It’s interesting to think that even if simSheer manages to sound like me, it will be doing so with no sense of aboutness at all – well, I *think* it will be doing so with no sense of aboutness. It has no senses, and no other data to tie my writings in with, so I don’t think that any of the neurons in it can possibly be tagged with any real world meaning. Or can they? This is probably a subject that some famous philosopher has held forth on and I should probably go try and find their works and read them, but in the meantime it’s certainly fun to think about.

I really wonder to what extent the aboutness problem (borrowed from Stephenson’s Anathem) applies to NNNs. Would the cluster I have for the concept of love even remotely resemble the clusters other people have? What would the differences say about me and them?

If this were a co-op game..

September 5th, 2017

So, I was having a conversation with a friend about one of my potential many mental models for God – the one in which God is a few neurons in each mind, spread out over all of us like a application running on a Beowulf cluster. In this particular model for God, it is possible that how we decide who goes to heaven and who goes to hell is majority vote. I hope this isn’t actually what’s going on, but you get some interesting results if it is.

Everyone goes to hell. Well, more likely, we throw religion out completely as criteria once we realize that everyone goes to hell.

Looking at a list of religions by population, you will see that *no one* has a majority vote. The top dog only has 31%. Now I can’t wrap my head around, at all, how people could be so dense as to think God is filtering based on religion and can’t manage to get the message straight. I’m not really all that clear on why anyone thinks God would need help multicasting a message while they believe God is all-powerful, but it strikes me that if we were playing a co-op game we’d be losing.

I do notice increasingly that men of faith are willing to admit that men of other faith are probably not evil nor the enemy. This is progress, but I think there’s a lot more to be made here. I can’t figure out how people even manage to hold the idea in their heads that A: our dispensation in the afterlife is limited to two destinations, given how big the universe obviously is B: there’s a omniscient deity who nonetheless can’t even manage to get a message to 1/3rd of the humans out there

Of course, this brings up the other (scary) possibility that not agreeing with the group you were born in is cause to be tormented for all eternity. But I would like to think that *no one* is going to be tormented for all eternity, because that phrase conjures up the idea of a being of pure, true evil. And yet, I do not get the feeling that the vast majority of Earth agrees with me that this is outside the realm of beleivability. This brings the idea of Peirson’s Puppiteers that the majority is always sane sharply into doubt.

I continue to want someone to author a new religion that doesn’t suck. What Scientology should have been but clearly wasn’t. If the Scientologists were honest, they would have the best neuroscience and mind-state gear in the world, instead of a 50-year-old dubious technology based on a wheatsone bridge and called a “e-meter”.

I do think the idea of why we can’t play Earth as a co-op game deserves further study

Cover: Sitting On The Dock Of The Bay

September 4th, 2017

Little blues – thank you Otis Redding..

Sitting On The Dock Of The Bay.

Neural networks and what you can’t let go of

August 9th, 2017

I had a interesting thought the other day about natural neural networks and people who hold beliefs that are not reality-verifiable or are even likely to be false. This thought started in looking at climate change deniers and people who believe religions that don’t appear to match the reality I’m experiencing, but it’s gone a bit further than that.

This is more of my hand-wavy guesswork.

It has occurred to me that one of the major problems a NNN faces is that subnets will tend to build major nexus points. These nexus points would appear to us to be core beliefs – or even just important beliefs. Once one of these beliefs is built, and a whole lot of connections to a whole lot of other subnets route through it, we would naturally be extremely resistant to removing it because we literally would be less able to function without it. In the case of religious (or religiously political) people – and I probably fit into this somewhat – letting go of their religion would make it far more difficult for their mind to work for a while – it would be somewhat similar to having a stroke. Major confluences of subnets which represented key ideas would no longer be valid – and it would likely be difficult to remove all of the traces of subnets like these, especially since there is a lot of redundancy in the way NNNs tend to wire. We may be extremely resistant to throw out cherished ideas – even when they’re proven wrong – because throwing them out makes it difficult for us to function at all, because all sorts of traffic is routed through them. They end up forming the underpinning for our personalities and decision trees.

I think if this is true, this is something we all need to understand and figure out the implications of. Christians brag of their faith being unshakable – but it might well be if Jesus showed up in person and told them they were wrong they would not be able to accept or integrate it because their faith is often loaded virally on them when they’re very young and ends up forming the physical underpinning for large portions of their mental structure.

Chester

July 22nd, 2017

So, I never met Chester. We had a lot of friends in common, for reasons that would take some explaining and probably aren’t worth going into here, and I am curious what he would have made of me if he had met me. But mostly, I feel a certain kinship to him, since we both have wrestled with some of the same demons. And thusly, I have written a song…

Chester

Godspeed and good luck, wherever you’re off to next.

Also available, Bunne’s remix: Chester, remixed by Bunne.