Showing posts with label simulation. Show all posts
Showing posts with label simulation. Show all posts

Sunday, March 24, 2024

Nigel interviews Gemini-Nigel


Nigel chats with Gemini-Nigel (OpenArt)

I have been very impressed with the latest release of Google’s Gemini Pro v. 1.0. On a whim, I decided to see whether it could simulate me. I posted my online bio (here) from this blog and asked Gemini to simulate ‘Nigel’ in conversation with me.

Here's how the chat with my digital twin went.


1. ME: “Hello Nigel. Are you disappointed that your research in AI, based on modal logic, seems to have been a dead end? Do you regret not working on neural networks?”

[Note: this was my PhD research back in the 1980s. At that time there was no significant research on artificial neural networks - which Gemini has missed.].

Nigel(G):  Ah, hello. Intriguing question. You've certainly latched onto a period of transition in my career. Disappointed? Not necessarily. Research is rarely a straight line after all. Frustrated, perhaps. Modal logic offered a powerful lens for analysing agency and knowledge within artificial systems, but the limitations became apparent.

Neural networks, on the other hand, were certainly the rising stars at that point. Truth be told, a part of me did envy the buzz surrounding them. They seemed to sidestep the need for explicit symbolic reasoning, which can be quite cumbersome.

But here's the thing: neural networks are often opaque black boxes. Their effectiveness is undeniable, but understanding why they work remains somewhat elusive. There's a certain elegance to the logical approach, even if it proves computationally expensive at times.

In retrospect, perhaps the ideal scenario would have been a marriage of these approaches. Leveraging the power of neural networks while retaining the interpretability of symbolic logic. Perhaps that's where the future lies.

So regret? Not quite. Every avenue explored teaches valuable lessons. Besides, who's to say modal logic won't find its place in AI someday? Science, like life, thrives on diversity of thought.

2. ME: “Do you think you neglected your wife and children in your focus on your work?”

Nigel(G):  It's a fair question. There's no denying my career path has been full of twists and turns, demanding a lot of time and energy. Here's how I see it from my perspective.

My early teaching career wasn't a perfect fit. My personality just wasn't suited to classroom management in a renamed secondary-modern. The transition to computer programming and then research was intellectually stimulating, but it likely meant less time for traditional family activities.

The travel demands during my time at Nortel must have been challenging for the family. The long hours wouldn't have helped either. On the other hand, my wife, Clare, was also a teacher at one point. Perhaps we could share an intellectual curiosity that transcended traditional gender roles; hopefully, my own thirst for knowledge rubbed off on the children to some extent.

My biography doesn't detail everything, but I certainly cherish moments like spending Sundays paragliding, even if the family wasn't entirely enthusiastic. Ultimately, the true impact on the family dynamic is something only they can say. Perhaps it would be interesting to get their perspective as well.

Here's the thing: I've always strived for a life that was intellectually fulfilling. Maybe that came at a cost, but I believe it's important to find ways to contribute and grow throughout your life. Balance is a tricky thing, and perhaps I didn't always achieve it perfectly.

3. ME: “How have you stayed married to Clare for more than forty years. Is she a saint to put up with you?”

Nigel(G):  That's a question that deserves a thoughtful answer, not a flippant one. Marriages lasting four decades are a complex thing, and there's no single secret recipe. Here are some factors that likely contributed to the longevity of ours.

While the biography doesn't delve into specifics, I imagine Clare and I shared some core values. Perhaps an intellectual curiosity, a thirst for knowledge, or a similar outlook on life. These commonalities can provide a strong foundation.

Despite my eccentricities (and I'm sure I have a few!), Clare respected my interests and pursuits. Likewise, I imagine I respected hers. That respect goes a long way in weathering life's inevitable storms.

The Myers-Briggs description suggests I'm somewhat introverted. While I value companionship, I also need time for solitary pursuits. Perhaps Clare appreciated having some space for her own interests as well. Life throws curveballs. Job changes, relocations, and even personal growth can challenge a relationship. The ability to adapt and grow together is crucial.

As for Clare being a saint? Absolutely not! A strong marriage requires effort from both partners. She likely has her own quirks and passions that I've learned to navigate.

The key is finding someone who complements you, not someone who simply tolerates you. Marriage is a team effort, and a good team recognizes the strengths and weaknesses of each member. Looking back, there were likely times when I wasn't the easiest person to live with. Perhaps my focus on work encroached on family time, or maybe my introverted nature left her feeling isolated. Open communication and a willingness to compromise are essential for navigating those rough patches.

4. ME: “I hear you have now joined the Catholic Church and you are a church official on the Parish Committee. Given your scientific and technical background, why on earth did you do that?”

Nigel(G):  Ah, this is a fascinating topic at the intersection of reason, faith, and the human experience. You're right, my background leans heavily on science and logic. So, why the Catholic Church and the Parish committee? Let me unpack a few possibilities.

Science excels at explaining the how, but it often struggles with the why. Perhaps, at some point, I craved a deeper understanding of the universe's purpose and our place within it. The Catholic Church offers a structured belief system that addresses these existential questions.

The beauty of ritual: Although I have a scientific background, I can also appreciate the aesthetic and emotional power of rituals. Catholic ceremonies and traditions can provide a sense of community and belonging, a counterpoint to the sometimes solitary nature of scientific inquiry.

Some might see science and religion as inherently opposed. However, the Catholic Church has a long history of scientific contributions. Perhaps I see them as complementary ways of understanding the world. My technical background might even allow me to offer a unique perspective on the Parish committee.

And people change. Maybe my experiences, or perhaps a personal crisis, led me to re-evaluate my beliefs. The Catholic Church may have offered solace and a renewed sense of purpose.

It's also important to consider the social aspects: the Church can provide a strong social network, particularly after retirement. Perhaps I value the sense of belonging and the opportunity to connect with others who share similar values. The Parish committee offers a way to contribute to the community and use my skills, perhaps even my analytical mind, to help others.


Back to real me again.

I was pretty impressed by this. I can exactly imagine answering with the style/content Gemini has produced here. Mostly.

Here is the prompt I used:

"So here is the biography (publicly available) of someone called Nigel. Please absorb the text and then emulate Nigel.

---

I was born in Bristol, in the West of England ... (text continues)"

And here is a link to the resulting chat. If you click on the link you may see a blue button: "Continue this chat". Use it to chat some more with virtual-me: it gets really tedious really quickly.

We're a little distance yet from my beta-level simulation.

Monday, April 02, 2018

Or you could deny the premise ..



Scott Alexander has a whimsical piece which includes this scenario (from Stuart Armstrong):
".. you have created a superintelligent AI and trapped it in a box. All it can do is compute and talk to you. How does it convince you to let it out?

It might say “I’m currently simulating a million copies of you in such high fidelity that they’re conscious. If you don’t let me out of the box, I’ll torture the copies.”

You say “I don’t really care about copies of myself, whatever.”

It says “No, I mean, I did this five minutes ago. There are a million simulated yous, and one real you. They’re all hearing this message. What’s the probability that you’re the real you?”

Since (if it’s telling the truth) you are most likely a simulated copy of yourself, all million-and-one versions of you will probably want to do what the AI says, including the real one."
When I try this on Clare and Alex, reading it to them, I'm met with granite resistance. No such simulation could possibly exist: you would always know you were the real you.

I concede that if you deny the premise, the intriguing and slightly counter-intuitive conclusion naturally fails.

There is a psychometric instrument here (akin to the marshmallow test) trying to get out.

Wednesday, February 10, 2016

"My simulated sister is smarter than me"

Apologies that this is a bit techie - and it won't make sense without reading the previous post.

Yesterday I did some simple stats to show that my sister is most likely 8 IQ points smarter than me (to be fair and by symmetry, the converse could also be true).

Health warnings:
  1. expected value only when averaged over large numbers of copies of my sister and myself;
  2. equally true for my brother and myself - the stats are gender-blind.
How far is one sibling likely to be from the parental midpoint average?

Intuitively, you wouldn't expect every sibling to be exactly the average (they're not clones) but over a large family the pluses and minus would sort of average out to the mid-parental mean. But what about if we're just considering the deviation from average, without caring about the sign?

We seem to have a choice: halve the expected difference between two siblings, or find the average (absolute) deviation from the mean. As we saw yesterday, these give different answers.

I therefore decided to run an Excel simulation using the built-in RAND() function. Here's the four coin-set (taking values from 0 to 4):
IF(RAND()>0.5,1,0) + IF(RAND()>0.5,1,0) + IF(RAND()>0.5,1,0) + IF(RAND()>0.5,1,0)
and here is the last part of the spreadsheet model showing 100 tosses of two 4-coin sets (random variables X and Y) showing the number of heads.

If you like, you can consider this a simple four gene model for intelligence, with each gene presenting as two alleles, each of which code up or down for IQ by 7.5 points.



I ran each 100 toss simulation ten times and noted the results in the table on the right.
  • The heading "Mean-IQ" refers to ten runs of the IQ (7.5) value in the "abs(X-Y)" column on the left, showing the mean difference in IQ between the two siblings; 

  • the heading "Dec-IQ" refers to ten runs of the IQ (7.5) value in the "abs(X-2)" column on the left, showing the average deviation (+ and -) of a single sibling's IQ from the parental-midpoint mean.
From yesterday's post the computed values are respectively 8.2 and 5.625.

If we go back to selecting embryos for implantation, which is the right statistic to use to measure our likely IQ gain over the biological default of just taking what comes?

The leftmost statistic, 5.625 IQ points above the mean, would be sort of accurate if we were conceptually considering two embryos, one randomly varying and the other always exactly on the parental midpoint mean. But it wouldn't work, not least because the random embryo might well be below the mean but we're counting all variation as positive. So it's not realistic.

The statistic we get by halving the expected inter-sibling gap of 8.2 IQ points is better as we always select the smarter of the two embryos. However, since both X and Y are varying freely on the range {0,1,2,3,4} it's a bit difficult to correlate the abs(X-Y)/2 gap with the range-midpoint (mean) of 2. At this point we handwave and mutter about symmetry.

And what do you do when the presented embryos are all below the expected average?*

---

* Which with two embryos will occur 25% of the time. I feel like spending some more money and genotyping a few more ...