
Andrew Grizzle
6.3K posts

Andrew Grizzle
@drewgriz
I wouldn't say I've been missing it, Bob @drewgriz.bsky.social



Dallas’ Highland Park votes to leave Texas’ second-largest public transit system. Voters in Highland Park chose to leave Dallas Area Rapid Transit amid complaints that the transit service isn’t worth the cost. bit.ly/4urckLj



🚨BREAKING: Gas hits $5.00/Gallon in Texas.


If you design/remodel a home too idiosyncratically to your own tastes It will be extremely challenging to sell later on, even in the blazing hot San Francisco real estate market of 2026



Hi. Restaurant and coffee shop owner here. Almond milk runs 2.2x the cost of whole milk. Oat milk is 2.4x, macadamia is 4x and soy is ~2x. We charge more because it costs us more, not because all of us are engaged in a rightwing plot against people who prefer alternatives.

There's a quadrillion-dollar question at the heart of AI: Why are humans so much more sample efficient compared to LLM? There are three possible answers: 1. Architecture and hyperparameters (aka transformer vs whatever ‘algo’ cortical columns are implementing) 2. Learning rule (backprop vs whatever brain is doing) 3. Reward function @AdamMarblestone believes the answer is the reward function. ML likes to use pretty simple loss functions, like cross-entropy. These are easy to work with. But they might be too simple for sample-efficient learning. Adam thinks that, in humans, the large number of highly specialised cells in the ‘lizard brain’ might actually be encoding information for sophisticated loss functions, used for ‘training’ in the more sophisticated areas like the cortex and amygdala. Like: the human genome is barely 3 gigabytes (compare that to the TBs of parameters that encode frontier LLM weights). So how can it include all the information necessary to build highly intelligent learners? Well, if the key to sample-efficient learning resides in the loss function, even very complicated loss functions can still be expressed in a couple hundred lines of Python code.







The network needs more connections, especially North-South



Governor Abbott’s threat to strip $110 million in public safety funding from Houston is an attempt to bully our city for doing what is right, what is legal, and what upholds the Constitution. The premise that Houston must either repeal this ordinance or lose funding is a false choice designed to force compliance through fear. Those are not the only options. Houston has a clear legal path forward. I sent a letter to urge the Mayor and the City Attorney to immediately file a Temporary Restraining Order (TRO) to prevent the State from withdrawing funds and seek a declaratory judgment affirming that this ordinance is lawful. If we act now and file, the court will issue an opinion as early as today, providing City Council with critical guidance before any rushed decision is made regarding repeal. There is no funding that will be put at risk by seeking this emergency relief ahead of the State’s April 20th deadline. Houstonians deserve a city that will stand up for them, defend its laws, and protect its residents. I respectfully urge the Mayor to take immediate legal action.


A two-story building will become 28 in Jamaica! Thanks to the Jamaica Neighborhood Plan, 285 homes — including permanently affordable housing — will rise just blocks away from the E, F, J and Z trains + LIRR. Read more in @CrainsNewYork: crainsnewyork.com/real-estate/re…


Many like to talk fondly about Paris-scale density without fully understanding what they’re advocating for. It’s very dense. It doesn’t have parking. It’s more than an architectural style and a building height.


Why aren't young people having kids? Because not only are housing+education costs too high, but they believe they have to uproot their entire life if/when they have children We need to make it easier to have little kids in the city (because that's where young people live)







