Tech Tuesday – Finland Edition (08/01/17)

Oscar Gordon

A Navy Turbine Tech who learned to spin wrenches on old cars, Oscar has since been trained as an Engineer & Software Developer & now writes tools for other engineers. When not in his shop or at work, he can be found spending time with his family, gardening, hiking, kayaking, gaming, or whatever strikes his fancy & fits in the budget.

Related Post Roulette

10 Responses

  1. Michael Cain says:

    Robo1 goes to the same automated weapon place as Robo3.Report

  2. aaron david says:

    I too carry a folding knife most of the time. And like Finns, as they are named after my son.Report

  3. fillyjonk says:

    Bio3 – while I know there are medications that can *slow* the progress of Alzheimer’s, there’s still no real treatment for it. How awful to know, “I’m doomed to develop this disease that will rob me of my memories and what makes me, me.”

    I would not be surprised to see physician-assisted suicides be the outcome of positive results on the blood test….I tend to be pretty anti-suicide but I confess if I were in that situation of “Yes, you’re not wrong, your memory is starting to go and it will only get worse” I might want a quicker exit than Nature would grant me.

    (Also, if the person were still working, would they lose their job? What would happen with their health insurance? These are all things I worry about when some test for an untreatable degenerative condition comes out….)Report

  4. veronica d says:

    [Engr1] — Now that is cool.

    Myself, I’m wondering if they could learn from the human-controlled total ordering and try to map the experiments to some kind of Gaussian process that over time could “branch out into weird regions for the purposes of learning” — which is a big advantage of the Gaussian processes approach.

    Which is to say, I get why this would be a bad idea when tweaking a literal fusion reactor. Broad jumps in parameter space are obviously a bad idea when creating tiny suns. On the other hand, I can see using a similar technique for less explosively dramatic experiments, where bigger jumps through parameter space would be worth it.

    Anyway, cool cool cool. This optimization nerd is glowing.

    (Not literally, unless there is secretly a fusion reactor under my desk.)Report

    • Oscar Gordon in reply to veronica d says:

      It makes sense, right? If you have a pretty good idea of the problem space and the variables involved, and you can develop some kind of heuristic algorithm to process the data from each shot, why not let the computer do it’s thing.

      So yeah, it’s pretty friggin’ cool.

      Also, I can’t find it, but I’ve linked to Tri-Alpha in the past, a few years back. I’m excited to see they are still plugging away at this.Report

    • Perhaps they need to apply themselves to fission…

      On Monday, the utility consortium building two new fission power reactors in South Carolina announced they were abandoning the project. South Carolina regulators previously allowed costs associated with the project — $9B in spending to date — into the rate base. Estimates are that 18% of the typical South Carolina Electric & Gas customer’s bill is due to costs associated with the now-abandoned project.Report

  5. Joe Sal says:

    Good stuff Oscar, I really enjoy these.Report

  6. Road Scholar says:

    WWW3: I just had a flashback to the BC comic strip.Report