See a doctor. It's an illness like any other, and drugs will help treat it. If the first med you try doesn't work, try another. Different things work for different people.
Some people close to you won't understand at first. That's ok. Their only people, and it's hard for them to have sympathy for an illness they cant physically see.
Do at least 1 productive thing everyday. Congratulate yourself for it.
Interesting how deep learning is evolving. Agreed that adversarial models are a huge advancement; it will be interesting to see how they progress over the next couple of years.
Yeah, gan's are huge. Mimicry plays a huge part in human learning; especially in language learning, where native fluency is more or less defined as 'difficult to distinguish from the real thing'. I expect we'll see a lot more coming from adversarial models in the future.
Naturally, when the result of a scientific study is entangled emotionally with something so personal like intelligence, it is hard to look at the results without bias. Lots of people are going to have a problem with this study, because it literally makes 99.999% of the population feel inadequate.
One way to make a system like this one more efficient/cost effective would be to draw the CO2 from ocean surface water which has a much higher concentration than the atmosphere, depending where you are in the world.
Flue gases with a high CO2 content are better than cheap; people will actually pay you to get rid of them. And we have a lot of them (this is kind of the problem in the first place). If you look at something like the exhaust from an oxyfuel turbine, you can get 70/30 N2/CO2 with relatively high purity.
For this reason, I'm always skeptical of atmospheric capture projects. It's just got to be more efficient to capture from a mixture with 20% (or more) CO2 than from a mixture with 0.04%.
Actually if the CO2 absorbent is efficient it is not much harder to capture CO2 from the atmosphere than from flue gas. The key really is the CO2 flux rate which wind provides.
The real advantage of atmosphere capture is you can do the capture where the cost is lowest, not where the CO2 is emitted.
> The key really is the CO2 flux rate which wind provides.
Indeed. This brings us to the issue of volumes. A useful carbon capture plant needs to capture at least 100 000 tons of CO2 per year (then we need "just" ~ 10 000 such plants). This means each of these minimal plants has to process an air volume of 250 billion cubic meters of air per year. At an air speed of 1 m/s (you can't flow too fast or there is no time for reaction), and assuming a wildly optimistic 50% capture from the filtered air, you need a reaction area of over 15 000 square meters.
Mind you, that's the area of the membrane which processes air. Add the auxiliary stuff around it, you can add at least another factor of 100 to the area, so each of your 10 000 plants have to be 1.5x the maximum planned size of the Tesla Gigafactory. And we're being optimistic.
Some people close to you won't understand at first. That's ok. Their only people, and it's hard for them to have sympathy for an illness they cant physically see.
Do at least 1 productive thing everyday. Congratulate yourself for it.