@azonenberg@ioc.exchange2026-03-28 19:03:52
ReworkCTF boards came in. I can't fully populate them because the Digikey order with the last couple of parts I need isn't coming until Monday but I should be able to populate the back side whenever.
I can also do one of the soldermask challenges, some of the back side inner layer rework, and the front side unpopulated BGA site rework if I want.
But first I have some logistics to sort out WRT streaming and which cameras go where etc to actually record the work. So I guess…
@lornajane@indieweb.social2026-05-27 07:40:39
I’ve ordered a refurbed pixel 8a as a stopgap, and will just miss things when I’ve left it somewhere around the house I guess! Hopefully some better options will show up at some point.
Mind you I had the iPhone mini for years and useful notifications on that are not allowed, presumably because they ruin the aesthetic. That never stopped being annoying.
@cdarwin@c.im2026-03-22 05:15:21
In a strongly worded decision this week, a federal judge ordered that the Voice of America
— its mission to provide news for countries around the world largely shut down for the past year by the Trump administration
— come roaring back to life.
Whether or not that actually happens is anybody’s guess.
The government filed notice Thursday to appeal U.S. District Court Judge Royce C. Lamberth’s order two days earlier to put hundreds of VOA employees who have been on paid l…
@arXiv_physicschemph_bot@mastoxiv.page2026-03-27 08:19:37
Autotuning T-PaiNN: Enabling Data-Efficient GNN Interatomic Potential Development via Classical-to-Quantum Transfer Learning
Vivienne Pelletier, Vedant Bhat, Daniel J. Rivera, Steven A. Wilson, Christopher L. Muhich
https://arxiv.org/abs/2603.24752 https://arxiv.org/pdf/2603.24752 https://arxiv.org/html/2603.24752
arXiv:2603.24752v1 Announce Type: new
Abstract: Machine-learned interatomic potentials (MLIPs), particularly graph neural network (GNN)-based models, offer a promising route to achieving near-density functional theory (DFT) accuracy at significantly reduced computational cost. However, their practical deployment is often limited by the large volumes of expensive quantum mechanical training data required. In this work, we introduce a transfer learning framework, Transfer-PaiNN (T-PaiNN), that substantially improves the data efficiency of GNN-MLIPs by leveraging inexpensive classical force field data. The approach consists of pretraining a PaiNN MLIP architecture on large-scale datasets generated from classical molecular simulations, followed by fine-tuning (dubbed autotuning) using a comparatively small DFT dataset. We demonstrate the effectiveness of autotuning T-PaiNN on both gas-phase molecular systems (QM9 dataset) and condensed-phase liquid water. Across all cases, T-PaiNN significantly outperforms models trained solely on DFT data, achieving order-of-magnitude reductions in mean absolute error while accelerating training convergence. For example, using the QM9 data set, error reductions of up to 25 times are observed in low-data regimes, while liquid water simulations show improved predictions of energies, forces, and experimentally relevant properties such as density and diffusion. These gains arise from the model's ability to learn general features of the potential energy surface from extensive classical sampling, which are subsequently refined to quantum accuracy. Overall, this work establishes transfer learning from classical force fields as a practical and computationally efficient strategy for developing high-accuracy, data-efficient GNN interatomic potentials, enabling broader application of MLIPs to complex chemical systems.
toXiv_bot_toot
@Sustainable2050@mastodon.energy2026-04-22 05:22:35
Both solar and wind power now produce more of the world's electricity than nuclear.
More importantly, they're the growth engines, ready to finally make a dent in the amounts of coal and gas used for electricity. And in the emissions arising from that!
Chart by Dan Schroeder on Bluesky.
@chrisnelder@mastodon.energy2026-03-23 19:12:27
Hey America, your President just spent $1 billion of your money in order to NOT get more energy from offshore wind.
TotalEnergies accepts $1B offshore wind buyout, pivots to oil and gas in US 😠 https://www.utilitydive.com/news/trump-administra…
@bourgwick@heads.social2026-04-23 03:07:52
40 years ago tonight, the last full grateful dead gig at the berkeley community theater, though almost all the band members will return for other benefits. pretty boss 2nd set jam. spring ’86, show #19.
soundboard: https://archive.org/details/gd1986-04-22.sbd.miller.34767.fla…
@simon_brooke@mastodon.scot2026-03-23 16:19:48
“We have not heard from the people responsible. I want them to appear before the ICC and hear what on earth was in their mind when they ordered 120mm tank rounds to be fired into an ambulance. Justice is first of all bringing the light of attention into this crime and secondly seeing the persons responsible being accountable for their actions.”
#GazaGenocide
@cdarwin@c.im2026-03-24 13:48:35
An oil refinery fire near the Texas coast was put out Tuesday and a shelter-in-place order was lifted,
hours after a large explosion at the complex shot plumes of smoke into the air, officials said.
No one was injured in Monday’s explosion at the Valero refinery in Port Arthur,
about 90 miles (145 kilometers) east of Houston, Mayor Charlotte M. Moses said. She had urged residents in parts of the west side of the city to stay put.
“There’s been an explosion, yes, but we…
@chrisnelder@mastodon.energy2026-03-11 00:31:39
Global energy crisis is underway thanks to Trump's war on Iran. Gov'ts ordering measures to save petrol & natural gas.
Thailand orders civil servants to work from home https://www.reuters.com/business/energy/th
