directevolve

The Interventions Testing Program is the closest thing I know of along these lines.

“ The Interventions Testing Program (ITP) is a peer-reviewed program designed to identify agents that extend lifespan and healthspan in mice. Investigators at any university, institute, company, or other organization are invited to recommend interventions for testing by submitting an application before the February deadline each year. Testing is carried out in the genetically heterogeneous UM-HET3 mouse stock at three sites — the Jackson Laboratory, the University of Michigan, and the University of Texas Health Science Center at San Antonio.”

https://www.nia.nih.gov/research/dab/interventions-testing-p...

Low dose extended release Adderall. I feel completely myself. It helps me make conscious choices about transitions between tasks. I don’t get irritated by interruptions. When I transition, it’s easier to put down what I was doing before and take up the new task.

If you think you have ADHD, try getting evaluated by a clinician. It’s not hard to do. A good clinician can go over the range of behavioral and medication options and assess for comirbidities like autism.

I had no trouble running Ubuntu on my ASUS Vivobook 4 years ago.

Yep. Bad use of force by police is tragic, but it’s happened before and will happen again. This is federal support for murdering liberals.

For the next elections, we need to convince swing voters and inconsistent voters to turn out for Democrats next election. Not reprogram the terminally MAGA.

The next onion layer will be credulous Fox viewers. But I’m even less clear on how we do that.

And in this case, shooting her made the situation much more dangerous. It caused her dead foot to stomp the gas pedal, dangerously accelerating the car uncontrollably until it crashed on the side of the road.

Vaccine studies use a different experimental design known as “longitudinal,” meaning they follow people over time. This study did not do that. It’s still a valid design, just limited in what it tells us.

As there were no healthy controls or comparison with blinded radiologist performance, it is not possible to answer those questions using this study. The point was to evaluate AI sensitivity, which is useful. However, without healthy controls, it is not possible to determine specificity, which is a crucial statistic since most women do not have breast cancer.

You can now see a very clear picture of exactly where that cop was standing with respect to the car as he drew his gun on the front page of the New York Times. He was not in the path of the car. And there was no reason for him to walk up and shoot her. The safe thing to do in that situation would have been to let her drive off. If they had wanted to arrest her, they could have used her license plate to find her later.

Police were shouting contradictory instructions at her, including to move her car. Her car had hardly starting to move when the cop walked around the front of the car and leaned in the window to shoot her in the face. Her corpse’s foot was on the gas, so the car then accelerated, crashing further down the road. That cop looks like a murderer to me from the footage. Hopefully, he will face accountability in a fair trial.

I was just reinforcing that point as your comment was worded in a way that left room for doubt. Sorry if this came across as critical toward you or implying you held a different interpretation.

No, this design is not capable of showing this. It did not compare to unblinded humans, did not provide AI with the same data used to make the initial diagnosis, and did not include healthy controls. It measures sensivitity only.

There were no healthy controls, so they can only measure sensitivity, not specificity in this design.

In case you missed it, they weren't using AI to make these diagnoses.

The did NOT test radiologists. There were NO healthy controls. They evaluated AI false negative rate and used exclusively unblinded radiologists to grade the level of visibility and other features of the cancer.

Utility of the study is to evaluate potential AI sensitivity if used for mass fully automated screenings using mammography data. But says NOTHING about the CRUCIAL false positive rate (no healthy controls) and NOTHING about AI vs. human performance.

See my main comment elsewhere in this threat.

The point of this study is that it suggests that fully AI-automated mammography can currently deliver 70% sensitivity in detecting breast cancer using this model. It does not enable us to compare AI to unaided human performance. As this study did not include healthy controls, there is no false positive rate. The false positive rate is a crucial missing metric, since the vast majority of women do not have breast cancer.

In nearly half the false negatives from both the mammogram and DWI datasets, the cancer was categorized as occult by two breast radiologists, meaning the cancer was invisible to a trained eye. The AI model's non-occult false negative rate on the mammography data is 19.3%.

For that 19.3% figure, see Table 2: 68 non-occult in AI-missed cancer, 285 non-occult in AI-detected cancer.

This study did not compare the AI to a radiologist on a mixed set of healthy and cancer images.

Yes, CRISPR’s mentioned in the article. Someone should write a book about it to break down the history leading up to this accomplishment.

Well, CRISPR-Cas9 as a tool for genetic engineering was only invented in 2012. A 13-year translation timeline is not unusual, maybe even unusually fast. CAR-T cell therapy for cancer took 30 years from discovery to clinic. It took about 30 years to go from the early attempts to use engineered lipid nano particles for drug delivery to the first FDA-approved medication using them, doxil.

With CRISPR, it took a long time to figure out how to reliably edit just the gene you want and acceptably minimize off-target edits, including by delivering the therapeutic to just the organ affected and getting the dose and release right.

The public is understandably leery about experimental medical techniques. If they had killed this newborn child with CRISPR therapy, then it might have set created a backlash delaying translation of this technique for years, possibly decades.

In biomedicine, we’re always looking for therapies that approximate the level of precision control available in software. Unfortunately, it’s never more than an approximation, and our ability to measure and predict the size of that error is always limited. That is why the field moves slowly.

I think it’s worth addressing this with more nuance.

Companies can get a METHOD-OF-USE PATENT (MOU) on an old drug for a new INDICATION.

This gives them the exclusive right to LABEL AND MARKET that drug for that indication for a period of time.

I however, doctors can prescribe and pharmacists can substitute generics for the new indication, regardless of the MOU.

For a company to profit off an MOU, they strategically need to create a new FORMULATION. This is a new dose or delivery mechanism (extended release, topical, etc). A new formulation can be protected with conventional patents that go beyond an MOU.

With an MOU + patent on a new formulation, the company has a brand where they are the only ones allowed to make the new formulation and the only ones with a product approved to be marketed for the new indication.

Getting FDA approval for the new brand is not the main hurdle for the company. To get insurers to pay the premium they want over the cost of existing treatment options, they have to show it’s safer or more effective than those existing options. Otherwise insurance will block it.

In principle, this means that repurposing should only enable companies to profitably repurpose off patent off label applications if they can provide a real patient benefit.

Whether this is the best or most efficient way to promote this kind of innovation, or whether it works as well in practice as it would seem in theory, is a separate question.

That’s an understandable misreading of what she said. I appreciate that the rhetorical effect could feel like a sly way to slip in asymmetric gender standards about how to interpret divorce.

But I am a pedantic person who prefers to focus on the literal statements in text rather than the perceived underlying emotional current. So I’ll pedantically plod through what she actually said.

She’s dealing with two dimensions of divorce: who initiated it (husband, wife, or collaborative), and whether it was surprising or unsurprising.

That gives several possibilities, but she lists three. What unifies them is that they are all written from the perspective of the abstract woman undergoing the experience.

1. Woman initiated, surprise unspecified.

2. Collaborative, so assume unsurprising.

3. Man initiated, surprising (her situation).

She doesn’t claim this covers all possibilities. The point of that bit is just to emphasize that divorces are different, and to object to treating them as a genre for wellness AI slop.

Here is the original text containing that part so others can easily form their own opinion.

“I also object to the flattening of the contours of my particular divorce. There are really important distinctions between the experiences of women who initiate their own divorces versus women who come to a mutual agreement with their spouses to end the marriage versus women, like me, who are completely blindsided by their husbands’ decisions to suddenly end the marriage. All divorces do involve self-discovery and rebuilding your life, but the ways in which you begin down that path often involve dramatically different circumstances.”

The important point here is that you can become much more effective by cutting wasted activity.

Wasted motion and power saps your time, energy, and motivation. It obscures your vision. It creates unnecessary risk. It decreases your awareness and adaptability.

Examples from my own life: - Teach kids to self-assess using an explicit rubric, instead of delivering feedback yourself every step of the way.

- Move slower and use the minimum power possible when doing chores.

- Jog more slowly and stay in zone 2, where you feel like you can go forever.

- Determine the smallest software performance test scale that makes the issue you’re trying to resolve reliably apparent.

- Resolve as much uncertainty as you can about new software features through separately scripted spikes before attempting to orchestrate them in the full codebase.

- Invest passively in index funds.

- Strip down presentations and pause a few seconds after most sentences. That lets the message sink in, gives you space to decide if you need to say more, and gives the audience a chance to respond if you’re not addressing their confusion effectively.

A good depiction of the gritty realities and the meaning of material striving for the very poor in turn of the century farm life is the novel Independent People, by Halldor Laxness, an Icelandic nobel laureate.

This is a wildly disingenuous interpretation of that study.

“ Using transaction-level data on US congressional stock trades, we find that lawmakers who later ascend to leadership positions perform similarly to matched peers beforehand but outperform them by 47 percentage points annually after ascension. Leaders’ superior performance arises through two mechanisms. The political influence channel is reflected in higher returns when their party controls the chamber, sales of stocks preceding regulatory actions, and purchase of stocks whose firms receiving more government contracts and favorable party support on bills. The corporate access channel is reflected in stock trades that predict subsequent corporate news and greater returns on donor-owned or home-state firms.”

https://www.nber.org/papers/w34524

> Montreal probably has the most restrictive rent control in Canada but it's also seeing the strongest apartment development growth

A wonderful city like Montreal can drive enough demand for housing to overcome red tape, and still be building far far less than what would satisfy demand. A less attribute city with lower demand for housing may build less due to lower demand, despite having less red tape.

Trust the economists on this one.

If most successful entrepreneurs are from ordinary backgrounds, one explanation is that most entrepreneurs, successful or unsuccessful, are from ordinary backgrounds. Another is that wealthy families might be pushing kids into entrepreneurship who lack the talent or taste for it.

That’s a statement of fact, but not intrinsically a problem. Being alive changes your brain chemistry.

The study found the average effect was no different from placebo. It’s plausible, based on anecdotes like the one you’re responding to, that Prozac is harmful for some people, helpful for others, with the expected benefit close to zero. If so, it might be necessary to just try it and see if it’s obviously helping, keep taking if so, stop if not.

This isn’t the whole story. I’ve been driving the same Prius compact since 2015, and only I the last few years have I suddenly started experiencing these insanely bright headlights. The giant cars have been around the whole time. The brightest lights seem like they are on specific brands of new cars, although I’m terrible at recognizing car brands and obviously only experience this issue at night when I’m being blinded by headlights, so it’s not easy to figure out exactly which brands cause the issue.

Charity navigator quantifies overhead. EA tried to quantify impact. To understand the difference, consider two hypothetical charities. Charity A has $1 million/year in administrative costs, while charity B’s costs are only $500,00/year.

Based on this, charity navigator says charity A is lower-ranked than charity B.

Now imagine that charity A and B can each absorb up to $1 billion in additional funding to work on their respective missions. Charity A saves one life for every $1,000 it gets, while B saves one life for every $10,000 it gets.

Charity navigator wouldn’t even attempt to consider this difference in its evals. EA does.

These evals get complex, and the EA organizations focused on charity evals like this have sophisticated methods for trying to do this well.

This is historically inaccurate. EA’s origins are in charity evalutions to quantify the marginal impact of a donation. This motivated philosophical debate about how to operationalize “good,” and later became influential enough to have political impact. Obviously EAs were inspired by philosophical ideas or even were philosophers. But that is not the same as it being the downstream practice of a uniform set of pre-existing philosophical commitments.