I think you’ll be pleasantly surprised: There was an awkward transition post-Tony, but the new hosts have finally hit their stride.
Natalie/Gus/Hilde are far more serene presences than Tony, but they let the rest of the cast shine through. Helen and Derek in particular are the standout presenters right now, and while they don’t have the same manic/ADHD vibe, they’re both passionate and curious in a way that reflects Tony’s style and keeps the narrative intriguing.
Obligatory plug for Atuin, which is a Sqlite-based shell history tool. It logs shell commands alongside timestamps, the working directory, and the return value.
You can optionally sync your history to a server in encrypted form to keep a shared history across hosts. The server is extremely easy to self host.
As discussed in TFA, it’s easy to filter or scrub junk like `cd ~/Desktop` so it doesn’t pollute your history. You can also fuzzy search and toggle between commands run in your current session, commands run on your current host, commands run in your CWD, and commands run on all hosts.
It’s my single favorite piece of dev tooling, and has made my job + life far smoother and easier. Highly recommend.
Ooooh, I'll have to take it for a test drive next week. This looks good for personal use as well as keeping a fleet of production maintenance/monitoring instances sync'd.
If you’re not on the bandwagon yet, you should try incorporating some mushroom blocks to speed up the decomposition! It’ll also help to take the moisture retention you’re getting up an additional notch.
We’ve got a local grow block recycling program through the Central Texas Mycological Society. For your use case, all you’d need to do is bury the blocks in leaf litter with one long edge barely poking up above the surface. A combo of blue and pink oyster would probably serve you well, depending on the season.
Since you’re the only other person in this thread to mention it, I’m surprised the industry has been sleeping so hard on QuickJS as a scripting language runtime.
I think the lack of batteries-included stdlib is probably the biggest holdup, but JS can be much nicer than python for anything that can be decomposed into map/reduce/filter string munging.
I agree. For me, the major stumbling block is what interpreters ship with your typical Linux distro. Python and BASH are usually just... there. Everything else just takes more steps.
Lest anyone take this seriously, these assertions are confidently-misinformed, conspiracy-minded thinking.
No asthma medications whatsoever have utility as a chemical masking agent, nor are there any plausible mechanisms for that to happen.
Beta agonists (mostly clenbuterol) have been abused independently in the past as a way to cut weight in weightlifting/cycling/etc., since they theoretically provide a marginal boost to overall metabolism - but the effects are marginal. They're de facto useless as a general PED.
Widespread doping in high-level sports is absolutely commonplace, and it's very easy to not get caught - but asthma medications have absolutely nothing to do with that.
The article you gave, they only state the principle of PEDS evasion tactics, some used a fake dick when it's time to urinate, some used compound modified, but it can't possibly tell every single way that scientists found to avoid detection.
Froom is an actual athlete that got caught, that speak louder.
Beside i heard it too in completely unrelated sport circle, running (sprint) and boxing from athlètes competing.
I work in the ICU monitoring field, on the R&D team of a company with live systems at dozens of hospitals and multiple FDA approvals. We use extended Kalman filters (i.e. non-blackbox "ML") to estimate certain lab values of patients that are highly indicative of them crashing, based on live data from whatever set of monitors they're hooked up to - and it's highly robust.
What the authors of this paper are doing is throwing stuff at the wall to see if it works, and publishing results. That's not necessarily a bad thing at all, but I say this to underline that their results are not at all reflective of SOTA capabilities, and they're not doing much exploration of prior art.
If you want to draw the line between ML and not ML, I think you’ll have to put Kalman filters and linear regression on the non-ML side. You can put support vector machines and neural networks on the ML side.
In some sense the exact place you draw the distinction is arbitrary. You could try to characterize where the distinction is by saying that models with fewer parameters and lower complexity tend to be called “not ML”, and models with more parameters and higher complexity tend to be called “ML”.
When you say it’s “obviously machine learning”, how could that statement possibly be correct? There’s not even broad consensus here… so you don’t get to say that your definition is obviously correct.
There are pedagogical reasons why you’d include linear regression in a machine learning course. This is pretty clear to me—they have properties which are extremely important to the field of machine learning field, such as differentiability.
It is obviously machine learning. It is a machine learning algorithm. It is taught in machine learning classes. It is described as a machine learning algorithm literally everywhere. Here's wiki:
> Linear regression is also a type of machine learning algorithm, more specifically a supervised algorithm, that learns from the labelled datasets and maps the data points to the most optimized linear functions that can be used for prediction on new datasets
You can pretend it's not because it's not a sophisticated machine learning algorithm, but you are wrong.
After spending over a decade in both statistics and machine learning I'd say the only reason there isn't a "broad consensus" is because statisticians like to gate-keep, whether that's linear regression, Monte Carlo methods, or Kalman Filters.
Linear regression appears in pretty much every ML textbook. Can you confidently say, "this model that appears in every ML textbook is the only model in the ML textbook that isn't an ML model"?
Kalman Filters are like a continuous-state HMM. So why are HMMs considered ML and Kalman Filters not considered ML?
IMO it's an ego thing. They spent decades rigorously analyzing everything about linear models and here come these CS cowboys producing amazing results without any of the careful rigor that statisticians normally apply. It's difficult to argue against real results so the inflexible, hard-nosed statisticians just hang on to whatever they can.
That’s unfairly harsh to statisticians, IMO. You have two fields of study, statistics and ML. There’s a massive overlap. Gatekeeping? Practitioners from these two fields have different jargon and view things differently from each other.
“X is taught in books about subject Y” is a pretty weak argument. I could use it to argue that group theory is quantum mechanics. Scientists and mathematics aren’t fighting over who gets to own group theory—the scientists get to put group theory in their toolboxes and the mathematicians get to study it for itself. Same with ML and statistics. When you do ML, you need certain statistical techniques in your toolbox, so they get taught in your ML books.
If you want to draw a line you could say anything that’s not full AGI isn’t machine learning because of philosophical reasons.
But, other than that there’s the only clear line is when the programmer isn’t hard coding results which puts Linear regression over the ML line. I guess you could argue about supervised vs unsupervised algorithms, but that’s going to exclude a lot of what is generally described as ML.
Anything that can separate data apoints can rightly been seen as a "supervised machine learning classifier".
Todemystify the area, I literally introduce my intro to ML lecture by drawing a line on the board, give its equation y = 0.5 x on the backboard, reminding students that they already know this, and then explain how to use it as a spam filter by interpreting the points on either side of the line as good emails versus spam ones.
Linear regression is machine learning. At their core neural networks are just repeated linear regression + a non-linearity arranged in interesting ways. The key is that they can be trained to fit data using some optimization protocol (e.g. gradient descent). Just because linear regression has a closed form solution and is conceptually simple doesn't mean anything here.
Obviously LLM and LRMs are the most studied, but even the recent posts on here from anthropic show that without a few high probability entries in the k-top results, confabulations are difficult for transformers.
Obviously there are PAC Learning methods that target anomaly detection, but they are very different than even EKF + Mc
You will note in this paper that even highly weighted features exhibited low sensitivity.
While the industry may find some pathological cases that make the approach usable, autograd and the need for parallelism make the application of this papers methods to tiny variations to multivariate problem ambitious.
They also only trained on medical data. Part of the reason the foundation models do so well is that they encode verifiers from a huge corpus that invalidates the traditional bias variance tradeoffs from the early 90's papers.
But they are still selecting from the needles and don't have access to the hay in the haystack.
The following paper is really not related except it shows how compression exacerbates that problem.
Chaitin's constant encoding the Halting problem, and that it is normal and uncomputable is the extreme top end of computability, but relates to the compression idea.
EKFs have access to the computable reals, and while non-linear, KF and EKFs can be thought of linearization of the approximations as a lens.
If the diagnostic indicators were both ergodic and Markovian, this paper's approach would probably be fairly reliable.
But these efforts are really about finding a many to one reduction that works.
I am skeptical about it in this case for PAC ML, but perhaps they will find a pathological case.
But the tradeoffs between statistical learning and expansive methods are quite different.
Obviously hype cycles drive efforts, I encourage you to look at this years AAAI conference report and see that you are not alone with the frustration on the single minded approach.
IMHO this paper is a net positive, showing that we are moving from a broad exploration to targeted applications.
A linear regression model can be written and trained as a neural net, has a loss function, all of that. Most if not all ML problems can be formulated as modelling a probability distribution
That’s too reductive—ML models are statistical models. Statistical models have parameters, and in general cases, you choose the parameters with some kind of optimization algorithm.
If you play fast and loose with your definition of “ML”, you’ll end up defining it so that any statistical model is an ML model… in which case, why even bother using two different terms?
ML models are, broadly speaking, the more complicated ones with more parameters, where the behavior of the models is not really known without training. I’m sure you could nitpick to death any definition I give, but that’s fine.
I am sure there are people teach data science classes who look at it in that "reductive" way.
From the viewpoint of engineering, scikit-learn provides the same interface to linear regression that it supplies to many other models. Huggingface provides an interface to models that is similar in a lot of ways but I think a 'regression' in that it doesn't provide the bare minimum of model selection facilities needed to reliably make calibrated models. There are many problems where you could use either linear regression or a much more complex model. When it comes to "not known without training" I'm not sure how much of that is the limit of what we know right now and how much is fundamental as in the problem of "we can't really know what a computer program with free loops with do" (Halting problem) or "we can't predict what side of the side Pluto is going to be on in 30 million years" (Deterministic chaos)
(The first industrial model trainer I built was a simultaneously over and under engineered mess like most things in this industry... I didn't appreciate scikit-learn's model selection facilities and even though the data sci's I worked with had a book understanding of them, they didn't really put them to work.)
There’s a pedagogical reason to teach things with a kind of reductive definition. It makes a lot of sense.
I remember getting cornered by somebody in a statistics class and interrogated about whether I thought neural networks were statistical techniques. In that situation I’ll only answer yes, they are statistical techniques. As far as I can tell, a big chunk of what we do with machine learning is create complicated models with a large number of parameters. We’re not creating something other than statistical models. We just draw a kind of cultural line between traditional statistics and machine learning techniques.
Now that I think about it, maybe if you asked me about the line between traditional statistics and machine learning, I would say that in traditional statistics, you can understand the parameters.
> in traditional statistics, you can understand the parameters.
I also think that this is the key differentiator between ML and stats.
Statistical models can be understood formally, which means that not only we know how each parameter affects the predictions, we also know what their estimation uncertainties are, under which assumptions, and how to check that these assumptions are satisfied. Usually, we value these models not only because they're predictive but also because they're interpretable.
In ML there is neither the luxury nor the interest in doing this, all we want is something that predicts as well as possible.
So the difference is not the model itself but what you want to get out of it.
I've heard professors call ML a form of applied statistics, and I think it's fair to call ML a subfield of statistics that deals with automatically generating statistical models with computers.
I'm questioning the association of a single maximum likelihood parameter estimation via analytical optimization as ML, because it does not invole any methods beyond calculus and no models other than the system itself, whose parameters we are estimating.
Perhaps I'm wrong, but the power of NN is in an unknown intermediate representation between the data (measurements in estimation) and the prediction. EKF has no such black box.
Which now that I've written it agrees with top level comment so I rescind my question.
I’ve emailed Dan and received the same response. My suspicions completely mirror yours.
I find this line particularly weak:
> We can only guess why users flag things
It’s not that hard to do some clustering analysis to see if bad-faith actors are repeatedly flagging posts in a coordinated manner. Maybe he’s trying to avoid giving away anti-spam secret sauce, but that doesn’t seem likely given the language of the copypasta. Speaking for myself, I would like any sort of assurance that anything other than a 100% laissez-faire approach to flag abuse is happening.
Natalie/Gus/Hilde are far more serene presences than Tony, but they let the rest of the cast shine through. Helen and Derek in particular are the standout presenters right now, and while they don’t have the same manic/ADHD vibe, they’re both passionate and curious in a way that reflects Tony’s style and keeps the narrative intriguing.
reply