n-e-w

“The data sets were randomly divided into training (85%) and test (15%) sets. We used 10-fold cross-validation to obtain generalized results of model performance. Data splitting was performed at the participant level and stratified based on the outcome variables. Because the data classes were imbalanced for symptom severity (ADOS-2 and SRS-2), we performed a random undersampling of the data at the participant level before conducting data splitting. Moreover, we examined different split ratios (80:20 and 90:10) to assess the robustness and consistency of the predictive performances across diverse splitting proportions.”

* undersampling is problematic here and probably introduced some bias. These imbalanced class problems are just plain hard. Claiming one hundred percent on an imbalanced class problem should probably cause some concern. * data split at the participant level has to be done really careful or you’ll over fit * multiple comparisons bias by testing multiple split ratios on the same test data. Same with the 10-fold cross Val. * not sure if they validated results on any external test data * outcome variable stratification also has to be done really carefully or it will introduce bias; seems particularly sensitive in this case * using severity of symptoms as class labels is problematic. These have to really have been diagnosed the same way / consistently to be meaningful.

I also note a long time history in collection of these images (15 years iirc). Hard to believe such a diverse set of images (collection, equipment etc) led to perfect results.

ML issues aside, super interested in the basic medical concept. I wasn’t aware retinal abnormalities could be indicative of issues like ASD.

It's not 1920. The modern middle manager has more invested in his fantasy football league than his direct reports -- and by some magnitude. Grow up.

[1] https://www.cell.com/iscience/fulltext/S2589-0042(22)01141-5...

I would say in 2018-2019 the ML / Deep learning stack was actually very impressive compared with what was available in FOSS. Now -- it's languishing. Badly.

It really is an astounding system. If you enjoy functional programming, the language is super expressive and productive. The integrated libraries are fantastic -- especially if you want to chain together different knowledge domains through a consistent interface and syntax. You can get the entire Wolfram Alpha knowledgebase and curated datasets to use in your programs immediately.

The front-end is just lightyears in front of Jupyter notebooks in just about every conceivable way.

This isn't to hide that there are some frustrating things about the language (like, machine learning development has just stopped dead after a couple of really great years of feature dev...why?) and, especially, the organisation and some of its personalities. But the product -- it really should get a lot more kudos and wow factor based on real accomplishments and features that are live, in the real world.

I love Wolfram stuff.

For non-subscribers

Hospitals are glorified hotels with super over-priced roomservice and activities. But the business model is very similar.

Less well known is uric acid's effect on hypertension -- which would also perhaps (anecdotally) explain likely elevated levels of heart attack and stroke in that same demographic.

All terribly confounded, though. Hard to unravel chickens from eggs, lifestyle factors etc.

I’ve had a non-trivial “addiction” to Diet Coke in various times of my life. MDs and Neuroscientists please weigh in! When I put my Bayesian hat on, I think this claim falls over (lots of people in general drink Diet Coke so the background numbers are high and people are likely to self-select as being “more intelligent” etc etc)

This observation is absolutely spot on. An ignored element of meritocracy is visibility -- and, perhaps, as you suggest, inertia. And then we end up back at the black hole of designing the 'right' incentive schemes...

sigh

Compassionate capitalism is more needed today than ever. Sure, this is likely just some low-level buffoon at a chain mouthing off in ignorance. But that's why it's dangerous; toxic capitalism has thoughtlessly metastasized to the middle.

Given the severe and extreme states most commodities markets have found themselves in of late, I expect an exacerbating withdrawal of liquidity by most market participants -- even posting margin for normal hedging requirements is beginning to require completely untenable amounts of free cash flow. I fear that severe inflation is very much looming at the door.

And that kind of hustle is a prerequisite to solid business success anyway. So, get after it!

Now, on with the first two.

"Hiring the best you can afford" doesn't mean hiring the best brand (Skadden. PwC etc). Find the person with the best technical expertise relevant to what you're pursuing. They're often stars from those shops who've gone out on their own -- just like you did.

Why? Don't you want to spend your money on a star engineer or great marketing?

No. Because you're in this game to win. And winning these days (whether going for an exit or, you know, an actual profitable business) needs you to have at least the following:

(1) Not make (or, more likely, not know how to avoid) stupid mistakes or ill-advised moves legally. (2) Not screw up the administration of your business

Even if you do everything else right, errors in either 1 or 2 can prove fatal. Like, literally, you will lose everything.

Good lawyers are like good soldiers these days. You don't go to war without soldiers. And you don't do modern business without the best legal help you can get. Even in the most advantageous, optimistic, friendly scenario, run it by an attorney who knows their stuff. First. During. Last. Or wander onto the battlefield with people who outgun you. You've been warned.

Next: good business administrators are worth their weight in gold. It is a distinct skill and some people really are called to love and be good at book-keeping, accounting, cash flows etc. Do not screw yourself on taxes, audits, book-keeping, invoices slipping through cracks, agreements not being fulfilled or anything in the "admin of business" that keeps you from staying laser focused on you do well -- which is not business admin.

Finally -- remember, business is business. Repeat it to yourself early and often. Business is business. The VC who seemed really friendly and just "gets you"? Business is business. That particular contract term that just doesn't sit right with you -- but the person on the other side keeps telling you not to worry about it? Business is business. Your cofounder is your best friend but now they're not pulling their quota? Business is business. This isn't being machiavellian. It's that everyone will be applying it to you.

Ultimately, good lawyers and good business administrators will ensure that you remember "business is business". Don't leave things to chance. Don't trust. Don't be emotional . Business is business and if you're not controlling your agenda, someone will be controlling it for you.

There's enough sh*t in business you can't control. At least stay in control of what you can.