Tag Archives: AI

What does “Neuromancer” have to teach us about the role of AI in society?

Posted on by
18

This post is by Phil Price, not Andrew.

From junior high through about sophomore year in college I read a lot of science fiction, went to some science fiction conventions, etc., but then I drifted away from the genre for eight or nine years. What brought me back was “Neuromancer”, by William Gibson. It had come out in 1984 when I was in college but I guess I had already stopped reading science fiction by then, or else I somehow missed that specific book, so I didn’t get around to reading it until about 1992. The book is generally credited as starting the “cyberpunk” sub-genre, of which Neal Stephenson’s “Snow Crash” is another great example, although there are many progenitors with similar DNA; indeed I’m not sure why the much movie Blade Runner isn’t given the credit (or perhaps something even earlier).

I haven’t read Neuromancer in about thirty years, but came across it while browsing a bookstore and thought eh, why not give it another read, the development of artificial intelligence is a major theme and maybe it’ll be interesting in that context, not just for entertainment.

There are some minor spoilers below, nothing that I think would interfere with one’s enjoyment of the book but if you are especially picky about this kind of thing then you might want to stop reading now.

In Neuromancer we encounter two types of artificial intelligence: artificial _general_ intelligence, as personified (machinified?) by AI’s known as Wintermute and Neuromancer; and ‘constructs’ such as the “Dixie Flatline construct”, which, we are told, is not truly intelligent but merely seems intelligent. The Dixie Flatline construct is “just a bunch of ROM” that answers questions the way a guy called “Dixie Flatline” would answer them himself. But then it turns out it’s not just about answering questions, Dixie Flatline can also hack into computer systems, pretty much like the person on whom it is based. And it can’t _just_ be ROM because it can remember things that you tell it.

I recall being somewhat puzzled by the distinction between the real AI’s and the “construct”, back when I read the book, since the construct sure _seems_ intelligent. But now that distinction seems entirely reasonable: the construct behaves very much like an LLM like chatGPT, which…well, I know there are people who think that as LLM’s get more sophisticated they are going to turn into artificial general intelligences, but I don’t think that’s the case. Artificial general intelligence is possible, and an LLM might even be a key component to attaining it, but I don’t think any LLM, no matter how grand, will be enough on its own. My younger self was puzzled by the distinction between the construct and a “real” AI, but now it makes perfect sense to me! Just think of the construct as an LLM that is trained to respond like a specific real person.

Another somewhat-realistic-seeming element of the book is that there’s an organization, colloquially called the “Turing Cops”, that is tasked with preventing AI’s from becoming too powerful. There’s a fear that if an AI becomes powerful enough it could destroy humanity, or at least do terrible things. There’s a lot of current discussion about whether or how AI’s should be regulated, although at least for now I don’t think that discussion is focused on the capabilities so much as who can use them and how, whereas in the book the Turing Cops only care how smart they get.

So…what about the title of this post, what does the book have to teach us about the role of AI in society? Nothing. Or at least, nothing I can think of. It’s a work of fiction written forty years ago by someone who, by his admission, knew nothing about the technologies he was writing about. Nowadays we might say he was “vibe-writing” or something. There’s lots of nutty stuff and some plot holes.

I guess I’ll mention one more thing that is purely on the literary side. The main character of the book, a hacker/cracker named Case, is objectively a horrible person, as is his girlfriend Molly. They lie, cheat, steal, kill, get involved in a scheme that kills dozens of innocent people and show no remorse about it, etc. But I spent the whole book rooting for them! The story is told mostly from Case’s point of view, and I kind of adopted his view of the world. He’s not without a sense of emotion or a sense of morality, but while reading I found that I liked the people he liked, disliked the people he disliked, was appalled by the things he found appalling but unbothered by the things that didn’t bother him. I don’t really have a point here, I just find the phenomenon interesting.

This post is by Phil.

Which AI coding assistant should I be using?

Posted on by
51

This post is by Phil Price, not Andrew.

For more than a year I have been using chatGPT to help write code. Maybe “help write code” is an understatement: often chatGPT writes the code to my instructions. I almost always get much better code in much less time than of I wrote everything myself, but I also frequently run into frustrating problems in which chatGPT acts like a chowderhead. (Here I’m referring to the o4-mini-high flavor of chatGPT.)

One recent experience: I need to do an SQL query: Use Table A to find the ID numbers of all of the customers in a specific group, cross-reference with Table B to find the locations and electric meter numbers associated with those customers, refer to Table C to get the dates and times that the group was selected for some kind of action, and refer to Table D to get the electricity consumption for those meters at those times. The variables have different names in the different tables (e.g. ‘customer_id’ in one table is called ‘customer_number’ in another table). The result is a somewhat involved query, but not an _extremely_ complicated query, you just have to go through one link at a time and be a bit careful. I thought chatGPT would nail this easily but in the end I spent more time coercing chatGPT to do it than it would have taken to do it myself: it would propose a chunk of code; I would try it and it would fail because (for instance) the variable name matching wasn’t done for every one of the steps or some similar issue; I would point out the problem and ask it to try again and it would propose another chunk of code with some different problem; and so on. I even tried the “deep thinking” option, which took much longer but still produced buggy code.

I mentioned this to a friend and he asked if I have tried Claude Code, which is his favorite. This has some features that seem potentially quite useful, including that you can have it scan your whole project code base and use that as context. That sounds great in a way, but also rather scary: what’s to stop it from going off the reservation and reading files I don’t want it reading? To some readers this may seem paranoid: do I really think Anthropic is going to steal my credit card information or something? Other readers will think I’m not nearly cynical enough: of _course_ these companies are going to do all kinds of unethical stuff, maybe not as crude as stealing my credit card information but not necessarily a lot better than that either. OK, yes, chatGPT has significant flaws as a programming assistant, but at least it only knows what I tell it and I have complete control over that.

So…I’m somewhat dissatisfied with ChatGPT o4-mini-high, I’m scared of Claude Code…what else should I consider? There are quite a few coding assistants, are there any that can write good code without the risk that my information will be used in ways I don’t like?

This post is by Phill.


What happened to genetic algorithms?

Posted on by
15

Eight years ago in March of 2017, evolutionary algorithms seemed on track to become the AI paradigm, before being supplanted by the LLMs that we all know and love (tolerate?). OpenAI proposed that evolutionary strategies could replace–or at least supplement–reinforcement learning: they are simple to implement and scale well. Since these optimizers are population-based, they’re parallelizable and make minimal assumptions. Three months later, however, Google released “Attention is All You Need” and the transformer was born. It took OpenAI roughly a year to then develop GPT1 (and we know the rest).

For those unfamiliar, genetic algorithms fall within a category of optimization procedures called metaheuristics. Acting analogously to evolution, they simulate populations of candidate solutions, select and retain the best, and modify the survivors for the next generation. By design, these algorithms lack closed-form solutions and strong guarantees. That’s their whole gimmick, and it’s a double-edged sword. They can solve black-box problems with practically no assumptions, but guarantees for finding optimal solutions are limited, and convergence speeds are slow and only known in specific contexts. 

Also, the true umbrella term is not actually genetic algorithms but “evolutionary computation” (EC), comprising four historically distinct subfields (though the schools have blended together in recent years):

  1. Genetic algorithms (the most commonly known), 
  2. Evolutionary strategies,
  3. Evolutionary programming, and
  4. Genetic programming. 

Given this is an applied statistics blog, what’s the relevance? It turns out that model selection (or as ML practitioners call it, hyperparameter tuning), can be framed as an optimization problem. Instead of doing grid search, LASSO, or old-fashioned forward selection, we can also search the space of models using evolutionary algorithms. In difficult-to-traverse, discrete model spaces, that could mean the difference between success and no attempt at all. 

What about beyond model selection? From the perspective of expanding models continuously, any unmodeled selection event (even if done with clever metaheuristics) still artificially shrinks posterior intervals and inflates confidence. Not an issue since evolutionary algorithms can also be used for general model estimation! That opens the door to optimizing model parameters over complicated non-differentiable geometries. Technically, we could even try EC instead of OLS for linear models (although I don’t know why we would want to).

So why is it that they have fallen in disfavor? Are they in disfavor?

Thomas Bäck, a prominent researcher in EC, (and collaborators) reviewed the last thirty years of developments. They claim the main issues that have prevented adoption are:

  1. A “bestiary” of nature-inspired algorithms lacking fundamental grounding has made the field incohesive, and practitioners confused
  2. There is no unifying paradigm for designing algorithms, and benchmarks only apply to specific problems, making specific implementations idiosyncratic and hard to evaluate

On the other hand, we shouldn’t undersell developments like CMA-ES (covariance matrix adaptation evolutionary strategy–a bit of a mouthful). In simple terms, you adapt a shared covariance matrix to shape, scale, and correlate your Gaussian perturbations based on current best candidates and past generations. This shared strategy is in contrast to previous algorithms where each individual updated its own parameters. It represents the current state of the art, ranks highly on numerical benchmarks, and has had success in combustion feedback control, multi-cellular migration, and ultrasound imaging. But then again, the core algorithm hasn’t changed in the thirty years since its inception, at least according to Bäck.

So, there may be a more fundamental issue, for which we can speculate. Evolving things takes lots of computation, and when I say lots, I truly mean lots.

A paper by Yampolskiy called “Why We Do Not Evolve Software” has made a compelling case for why evolutionary algorithms aren’t a cure all. The following are incredibly rough estimates and make various assumptions about total global biomass, cell replication rates, and what constitutes biological computation, but essentially: the biosphere that evolved us can compute ~10^40+ FLOPS (floating point operations) per second. It also took that biosphere billions of years to evolve us. All our supercomputers on Earth combined can only compute ~10^22 FLOPS per second. This would seem to imply that the cleverness of an efficient genetic search has to cut through ten to thirty orders of magnitude! That’s a hefty requirement. Not impossible, but probably impractical in many cases.

What does this mean about the future of genetic algorithms? It seems that solutions will somehow need to bypass the combinatorial explosion of the solution space, since getting the compute required without that would take hundreds of years, even with an ever-constant Moore’s law. One direction may be to continue exploring the interplay between local and global solutions (like in CMA-ES and particle swarm optimization). The most concrete step forward would be improving benchmarks and validation. Maybe a unifying paradigm for automatically designing evolutionary algorithms could then develop, similar to what we have in probabilistic modeling.

Beat the bot in this year’s Cherry Blossom Prediction Competition

Posted on by
3

This post is by Lizzie.

It’s back with a twist!

Once again, the Cherry Blossom Prediction Competition will run throughout February 2025. We challenge you to predict the bloom date of cherry trees at five locations throughout the world and win prizes.

However, this year, contestants will not only compete against each other for the top prizes—but against artificial intelligence.

We will include one or more submissions from the most popular large language models. Our AI handlers will prompt the AI with the contest rules and the entries from previous competitions. The handlers will then execute any code written by the AI as part of their entry. Judges will review all entries without knowing which were submitted by humans and which were written by AI.

Any human that beats the AI will receive commemorative memorabilia indicating they “beat the bot in the 2025 International Cherry Blossom Prediction Competition.”

I’ve been mistaken for a chatbot

Posted on by
31

… Or not, according to what language is allowed.

At the start of the year I mentioned that I am on a bad roll with AI just now, and the start of that roll began in late November when I received reviews back on a paper. One reviewer sent in a 150 word review saying it was written by chatGPT. The editor echoed, “One reviewer asserted that the work was created with ChatGPT. I don’t know if this is the case, but I did find the writing style unusual ….” What exactly was unusual was not explained.

That was November 20th. By November 22nd my computer shows a file created named ‘tryingtoproveIamnotchatbot,’ which is just a txt where I pasted in the GitHub commits showing progress on the paper. I figured maybe this would prove to the editors that I did not submit any work by chatGPT.

I didn’t. There are many reasons for this. One is I don’t think that I should. Further, I suspect chatGPT is not so good at this (rather specific) subject and between me and my author team, I actually thought we were pretty good at this subject. And I had met with each of the authors to build the paper, its treatise, data and figures. We had a cool new meta-analysis of rootstock x scion experiments and a number of interesting points. Some of the points I might even call exciting, though I am biased. But, no matter, the paper was the product of lots of work and I was initially embarrassed, then gutted, about the reviews.

Once I was less embarrassed I started talking timidly about it. I called Andrew. I told folks in my lab. I got some fun replies. Undergrads in my lab (and others later) thought the review itself may have been written by chatGPT. Someone suggested I rewrite the paper with chatGPT and resubmit. Another that I just write back one line: I’m Bing.

What I took away from this was myriad, but I came up with a couple next steps. I decided this was not a great peer review process that I should reach out to the editor (and, as one co-author suggested, cc the editorial board). And another was to not be so mortified as to not talk about this.

What I took away from these steps were two things:

1) chatGPT could now control my language.

I connected with a senior editor on the journal. No one is a good position here, and the editor and reviewers are volunteering their time in a rapidly changing situation. I feel for them and for me and my co-authors. The editor and I tried to bridge our perspectives. It seems he could not have imagined that I or my co-authors would be so offended. And I could not have imagined that the journal already had a policy of allowing manuscripts to use chatGPT, as long as it was clearly stated.

I was also given some language changes to consider, so I might sound less like chatGPT to reviewers. These included some phrases I wrote in the manuscript (e.g. `the tyranny of terroir’). Huh. So where does that end? Say I start writing so I sound less to the editor and others ‘like chatGPT’ (and I never figured out what that means), then chatGPT digests that and then what? I adapt again? Do I eventually come back around to those phrases once they have rinsed out of the large language model?

2) Editors are shaping the language around chatGPT.

Motivated by a co-author’s suggestion, I wrote a short reflection which recently came out in a careers column. I much appreciate the journal recognizing this as an important topic and that they have editorial guidelines to follow for clear and consistent writing. But I was surprised by the concerns from the subeditors on my language. (I had no idea my language was such a problem!)

This problem was that I wrote: I’ve been mistaken for a chatbot (and similar language). The argument was that I had not been mistaken — my writing had been. The debate that ensued was fascinating. If I had been in a chatroom and this happened, then I could write `I’ve been mistaken for a chatbot’ but since my co-authors and I wrote this up and submitted it to a journal, it was not part of our identities. So I was over-reaching in my complaint. I started to wonder: if I could not say ‘I was mistaken for an AI bot’ — why does the chatbot get ‘to write’? I went down an existential hole, from which I have not fully recovered.

And since then I am still mostly existing there. On the upbeat side, writing the reflection was cathartic and the back and forth with the editors — who I know are just trying to their jobs too — gave me more perspectives and thoughts, however muddled. And my partner recently said to me, “perhaps one day it will be seen as a compliment to be mistaken for a chatbot, just not today!”

Also, since I don’t know an archive that takes such things so I will paste the original unedited version below.

I have just been accused of scientific fraud. It’s not data fraud (which, I guess, is a relief because my lab works hard at data transparency, data sharing and reproducibility). What I have just been accused of is writing fraud. This hurts, because—like many people—I find writing a paper a somewhat painful process.

Like some people, I comfort myself by reading books on how to write—both to be comforted by how much the authors of such books stress that writing is generally slow and difficult, and to find ways to improve my writing. My current writing strategy involves willing myself to write, multiple outlines, then a first draft, followed by much revising. I try to force this approach on my students, even though I know it is not easy, because I think it’s important we try to communicate well.

Imagine my surprise then when I received reviews back that declared a recently submitted paper of mine a chatGPT creation. One reviewer wrote that it was `obviously Chat GPT’ and the handling editor vaguely agreed, saying that they found `the writing style unusual.’ Surprise was just one emotion I had, so was shock, dismay and a flood of confusion and alarm. Given how much work goes into writing a paper, it was quite a hit to be accused of being a chatbot—especially in short order without any evidence, and given the efforts that accompany the writing of almost all my manuscripts.

I hadn’t written a word of the manuscript with chatGPT and I rapidly tried to think through how to prove my case. I could show my commits on GitHub (with commit messages including `finally writing!’ and `Another 25 mins of writing progress!’ that I never thought I would share), I could try to figure out how to compare the writing style of my pre-chatGPT papers on this topic to the current submission, maybe I could ask chatGPT if it thought I it wrote the paper…. But then I realized I would be spending my time trying to prove I am not a chatbot, which seemed a bad outcome to the whole situation. Eventually, like all mature adults, I decided what I most wanted to do was pick up my ball (manuscript) and march off the playground in a small fury. How dare they?

Before I did this, I decided to get some perspectives from others—researchers who work on data fraud, co-authors on the paper and colleagues, and I found most agreed with my alarm. One put it most succinctly to me: `All scientific criticism is admissible, but this is a different matter.’

I realized these reviews captured both something inherently broken about the peer review process and—more importantly to me—about how AI could corrupt science without even trying. We’re paranoid about AI taking over us weak humans and we’re trying to put in structures so it doesn’t. But we’re also trying to develop AI so it helps where it should, and maybe that will be writing parts of papers. Here, chatGPT was not part of my work and yet it had prejudiced the whole process simply by its existential presence in the world. I was at once annoyed at being mistaken for a chatbot and horrified that reviewers and editors were not more outraged at the idea that someone had submitted AI generated text.

So much of science is built on trust and faith in the scientific ethics and integrity of our colleagues. We mostly trust others did not fabricate their data, and I trust people do not (yet) write their papers or grants using large language models without telling me. I wouldn’t accuse someone of data fraud or p-hacking without some evidence, but a reviewer felt it was easy enough to accuse me of writing fraud. Indeed, the reviewer wrote, `It is obviously [a] Chat GPT creation, there is nothing wrong using help ….’ So it seems, perhaps, that they did not see this as a harsh accusation, and the editor thought nothing of passing it along and echoing it, but they had effectively accused me of lying and fraud in deliberately presenting AI generated text as my own. They also felt confident that they could discern my writing from AI—but they couldn’t.

We need to be able to call out fraud and misconduct in science. Currently, the costs to the people who call out data fraud seem too high to me, and the consequences for being caught too low (people should lose tenure for egregious data fraud in my book). But I am worried about a world in which a reviewer can casually declare my work AI-generated, and the editors and journal editor simply shuffle along the review and invite a resubmission if I so choose. It suggests not only a world in which the reviewers and editors have no faith in the scientific integrity of submitting authors—me—but also an acceptance of a world where ethics are negotiable. Such a world seems easy for chatGPT to corrupt without even trying—unless we raise our standards.

Side note: Don’t forget to submit your entry to the International Cherry Blossom Prediction Competition!

More than 10k scientific papers were retracted in 2023

Posted on by
6

Hi all, here to talk about one of my favorite scientific topic: integrity and correction of science.

Here comes some good news for most of us and of humanity. More than 10k scientific papers have been retracted this year. Aside from the researchers who have received these notices of retractions (some of them for multiple papers), and the publishers, this is quite good news I would argue. This comes after a big year on this topic and the topic of finding fraudulent practices (see, for instance, how Guillaume Cabanac easily found papers generated by ChatGPT) and very problematic journals with, for instance, Hindawi journals probably being more problematic than others. Many retractions and reports have focused on duplicated images or use of tortured phrases. New fraudulent practices have also emerged and been found (see for instance our findings about “sneaked references” that some editors/journals have manipulated the metadata of accepted papers to increase citations of specific scholars and journals).

Of course, some like me may always see the glass half empty and I would still argue that probably many more papers should have been retracted and that, as I have lamented many times, the process of correcting the scientific literature is too slow, too opaque, and too bureaucratic while at the same time not protecting, funding, or rewarding the hard-working sleuth behind the work. Most of the sleuthing work takes place in spite of, rather than thanks to, the present publication and editorial system. Often the data or metadata to facilitate investigations is not published or available (e.g., lack of metadata about ethics or lack of metadata about reviewing practices).

Still, I guess it is kind of victory that sleuthing work is taken seriously these days I suppose, and I would like to take the opportunity of this milestone of 10k retracted paper to invite some of you to also participated in Pubpeer discussions. I am sure your input would be quite helpful there.

Happy to read thoughts and comments on the milestone and its importance. I will continue to write (a bit more regularly I hope) here on this topic.

Lonni Besançon

 

 

 

 

 

 

“Unsupervised learning” gets a bad rap

Posted on by
17

Yesterday, a speaker from the company formerly known as Facebook corrected themselves after mentioning “unsupervised learning,” saying they weren’t supposed to use that term any more. This must be challenging after years of their chief AI scientist promoting unsupervised learning. The reason is apparently that they don’t want the public to worry about the rise of unsupervised AI! Instead, they are using the terms “unlabeled” and “self supervised”.

If I were the lexical police, I would’ve blacklisted the term “learning,” because we’re really just estimating parameters (aka “weights”) of a statistical model. I felt bad saying “learning” even back when I worked in ML full time, but that’s probably just because I lived through the AI winter during which the quickest route to rejection of a paper or grant was to mention “artificial intelligence.”

EU proposing to regulate the use of Bayesian estimation

Posted on by
45

The European Commission just released their Proposal for a Regulation on a European approach for Artificial Intelligence. They finally get around to a definition of “AI” on page 60 of the report (link above):

‘artificial intelligence system’ (AI system) means software that is developed with one or more of the techniques and approaches listed in Annex I and can, for a given set of human-defined objectives, generate outputs such as content, predictions, recommendations, or decisions influencing the environments they interact with

We don’t even have to wonder if they mean us. They do. Here’s the full text of Annex 1 from page 1 of “Laying down harmonized rules on artificial intelligence (Artificial Intelligence Act) and amending certain Union legislative acts” (same link, different doc).

ANNEX I
ARTIFICIAL INTELLIGENCE TECHNIQUES AND APPROACHES
referred to in Article 3, point 1

(a) Machine learning approaches, including supervised, unsupervised and reinforcement learning, using a wide variety of methods including deep learning;

(b) Logic- and knowledge-based approaches, including knowledge representation, inductive (logic) programming, knowledge bases, inference and deductive engines, (symbolic) reasoning and expert systems;

(c) Statistical approaches, Bayesian estimation, search and optimization methods.

This feels hopelessly vague with phrasing like “wide variety of methods” and the inclusion of “statistical approaches”. Also, I don’t see what’s added by “Bayesian estimation” given that it’s an instance of a statistical approach. At least it’s nice to be noticed.

Annex I looks like my CV rearranged. In the ’80s and ’90s, I worked on logic programming, linguistics, and knowledge representation (b). I’m surprised that’s still a going concern. I spent the ’00s working on ML-based natural language processing, speech recognition, and search (a). And ever since the ’10s, I’ve been working on Bayesian stats (c).

Is data science a discipline?

Posted on by
31

Jeannette Wing, director of the Columbia Data Science Institute, sent along this link to this featured story (their phrase) on their web site.

Is data science a discipline?

Data science is a field of study: one can get a degree in data science, get a job as a data scientist, and get funded to do data science research. But is data science a discipline, or will it evolve to be one, distinct from other disciplines? Here are a few meta-questions about data science as a discipline.

  • What is/are the driving deep question(s) of data science? Each scientific discipline (usually) has one or more “deep” questions that drive its research agenda: What is the origin of the universe (astrophysics)? What is the origin of life (biology)? What is computable (computer science)? Does data science inherit its deep questions from all its constituency disciplines or does it have its own unique ones?
  • What is the role of the domain in the field of data science? People (including this author) (Wing, J.M., Janeia, V.P., Kloefkorn, T., & Erickson, L.C. (2018)) have argued that data science is unique in that it is not just about methods, but about the use of those methods in the context of a domain—the domain of the data being collected and analyzed; the domain for which a question to be answered comes from collecting and analyzing the data. Is the inclusion of a domain inherent in defining the field of data science? If so, is the way it is included unique to data science?
  • What makes data science data science? Is there a problem unique to data science that one can convincingly argue would not be addressed or asked by any of its constituent disciplines, e.g., computer science and statistics?

I don’t understand how bullet point two is supposed to distinguish data science from the more prosaically titled field of applied statistics.

The story goes on to enumerate ten research challenges in data science. Some of them are hot AI topics like ethics and fairness, some of them are computer science topics such as computing systems for data-intensive applications, and some of them are statistics topics like causal inference.

Melanie Mitchell says, “As someone who has worked in A.I. for decades, I’ve witnessed the failure of similar predictions of imminent human-level A.I., and I’m certain these latest forecasts will fall short as well. “

Posted on by
37

Melanie Mitchell‘s piece, Artificial Intelligence Hits the Barrier of Meaning (NY Times behind limited paywall), is spot-on regarding the hype surrounding the current A.I. boom. It’s soon to come out in book length from FSG, so I suspect I’ll hear about it again in the New Yorker.

Like Professor Mitchell, I started my Ph.D. at the tail end of the first A.I. revolution. Remember, the one based on rule-based expert systems? I went to Edinburgh to study linguistics and natural language processing because it was strong in A.I., computer science theory, linguistics, and cognitive science.

On which natural language tasks can computers outperform or match humans? Search is good, because computers are fast and it’s a task at which humans aren’t so hot. That includes things like speech-based call routing in heterogeneous call centers (something I worked on at Bell Labs).

Then there’s spell checking. That’s fantastic. It leverages simple statistics about word frequency and typos/brainos and is way better than most humans at spelling. It’s the same algorithms that are used for speech recognition and RNA-seq alignment to the genome. These all sprung out of Claude Shannon’s 1948 paper, “A Mathematical Theory of Communication”, which has over 100K citations. It introduced, among other things, n-gram language models at the character and word level (still used for speech recognition and classification today with different estimators). As far as I know that paper contained the first posterior predictive checks—generating examples from the trained language models and comparing them to real language. David McKay’s info theory book (the only ML book I actually like) is a great introduction to this material and even BDA3 added a spell-checking example. But it’s hardly A.I. in the big “I” sense of “A.I.”.

Speech recognition has made tremendous strides (I worked on it at Bell Labs in the late 90s then at SpeechWorks in the early 00s), but its performance is still so far short of human levels as to make the difference more qualitative than quantitative, a point Mitchell makes in her essay. It would no more fool you into thinking it was human than an animatronic Disney character bolted to the floor. Unlike games like chess or go, it’s going to be hard to do better than people at language, but it would certainly be possible. But it would be hard to do that the same way they built, say Deep Blue, the IBM chess-playing hardware that evaluated so many gazillions of board positions per turn with very clever heuristics to prune search. That didn’t play chess like a human. If the better language was like that, humans wouldn’t understand it. IBM Watson (natural language Jeopardy playing computer) was closer to behaving like humans with its chain of associative reasoning—to me, that’s the closest we’ve gotten to something I’d call “A.I.”. It’s a shame IBM’s oversold it since then.

Human-level general purpose A.I. is going to be an incredibly tough nut to crack. I don’t see any reason it’s an unsurmounable goal. It’s not going to happen in a decade without a major breakthrough. Better classifiers just aren’t enough. People are very clever, insanely good at subtle chains of associative reasoning (though not so great at logic) and learning from limited examples (Andrew’s sister Susan Gelman, a professor at Michigan, studies concept learning by example). We’re also very contextually aware and focused, which allows us to go deep, but can cause us to miss the forest for the trees.

A.I. parity with the West in 2020

Posted on by
8

Someone just sent me a link to an editorial by Ken Church, in the journal Natural Language Engineering (who knew that journal was still going? I’d have thought open access would’ve killed it). The abstract of Church’s column says of China,

There is a bold government plan for AI with specific milestones for parity with the West in 2020, major breakthroughs by 2025 and the envy of the world by 2030.

Something about that plan sounded familiar. Then I remembered the Japanese Fifth Generation project. Here’s Ehud Shapiro, writing a trip report for ACM  35 years ago:

As part of Japan’s effort to become a leader in the computer industry, the Institute for New Generation Computer Technology has launched a revolutionary ten-year plan for the development of large computer systems which will be applicable to knowledge information processing systems. These Fifth Generation computers will be built around the concepts of logic programming. In order to refute the accusation that Japan exploits knowledge from abroad without contributing any of its own, this project will stimulate original research and will make its results available to the international research community.

My Ph.D. thesis, circa 1989, was partly on logic programming, as was my first book in 1992 (this post isn’t by Andrew, just in case you hadn’t noticed). Unfortunately, by the time my book came out, the field was pretty much dead, not that it had ever really been alive in the United States. As an example of how poorly it was regarded in the U.S., my first grant proposal to the U.S. National Science Foundation, circa 1990, was rejected with a review that literally said it was “too European.”