I'm thinking about a new way of interviewing candidates for a PhD. The idea would be to assume they're smart and competent, take their research proposal and do a 1h brainstorming session role-playing the first week of the PhD. Good idea? Could it introduce biases? Improvements?
@8599d6ab@5bafa990 no but you might wonder what it would have been like to continue doing well funded research rather than drowning in admin and teaching given to you because you weren't bringing in enough research income to the university. Sigh. Hasn't happened to me yet but I worry.
What neuroscience / comp neuro papers would you put on a recommended reading list if you wanted to emphasise the creativity, inspiration and joy of the field? I think some suggestions would overlap with the most famous or epochal papers, but some might be quite different.
#neuroscience #compneuro
@8599d6ab I also have ups and downs. Recently I've been trying to let go of things that are "good for my career" to focus on things I actually care about. It feels good, although I worry about the long term effect on my career of course.
Many thanks to all who sent abstracts for the SNUFA SNN workshop. We will be sending every participant a random sample of abstracts to vote on to help decide what should be a talk/poster. If you want to take part, register today (free).
https://www.eventbrite.co.uk/e/snufa-2023-tickets-675972952297
More info on the workshop at:
https://snufa.net/2023/
#SpikingNeuralNetworks #compneuro #snn #neuroscience
@8f5529d2@ab7cf35b I keep hearing how great it is and have been meaning to try it but not got around to it yet, so it seemed like a good example. 😂
I think we agree we just put it in different words. The "slow science" phrasing just kind of puts me off because it sounds like the opposite of what I want.
@ab7cf35b I want science to be fast. Why wouldn't I want discoveries to be made faster? I just don't want it to be fast because it's half baked bullshit to boost metrics.
This week I read about a Nobel winner whose groundbreaking work didn't get funded and got her demoted, and about data fraud by two of the highest profile scientists who were lauded and mega funded. We have to stop rewarding short term flashy work and overproductive scientists.
It's fine and correct to talk about both incentives and individual responsibility. But if we scientists collectively decided to heavily downplay work without open, raw data and reproducible methods, and ignored journal title when evaluating scientists, this couldn't happen.
The system is absolutely broken and needs structural reform, yes. Journals need to go. Competitive grants are the wrong way to fund science. Scientific prizes are very problematic. But we also need to get better at reading and doing science and valuing what works in the long term.
That's the key point. If we let these things happen it means we are doing science badly.
@5bafa990@8599d6ab no no no! That's not the way towards responsible well evidenced change. My 15 page statistical analysis shows that changing the publications weighting factor from 2 to 1.9 will lead to a 1.3% improvement in the rate of fundamental discoveries (with 63% confidence). Yes it'll take years of work to get this change implemented but it's worth it. That's change we can believe in.
@0553753f I found this article very frustrating because I also daily see this lack of solidarity and the terrible effects of it, but I don't think this article will persuade anyone to change their ways. It feels nihilistic. Better question to ask is why have we who believe in improving things via solidarity failed to convince everyone?
@e298e1ac let them spend the first 6 months mucking about trying whatever they like and that's good for them to explore their interests and gives you an idea of what they'd do best at.
The first book (1898) on what became reinforcement learning starts with pages of anger on how stupid animals are: "Dogs get lost hundreds of times and no one ever notices it or sends an account of it to a scientific magazine. But let one find his way from Brooklyn to Yonkers and the fact immediately becomes a circulating anecdote" 😂
Sometimes I wish I could have worked in that age.
Knock your socks off here, it's well worth a read:
https://archive.org/download/animalintelligen00thoruoft/animalintelligen00thoruoft.pdf
And don't imagine that cats get off lightly either. "Thousands of cats on thousands of occasions sit helplessly yowling, and no one takes thought of it or writes to his friend, the professor; but let one cat claw at the knob of a door supposedly as a signal to be let out, and straightway this cat becomes the representative of the cat-mind in all the books."
Current procrastination level: I just did surgery on a computer mouse. Took it apart and removed a load of fluff and it works again. Seems like the mouse wheel works by registering the passing of spokes with a tiny laser. Fascinating! Back to prepping teaching? Sigh.
@ab7cf35b check out my student's paper on a graph layout algorithm that is particularly easily adaptable to add additional constraints. Comes with software. 😉
http://neural-reckoning.org/pub_graph_drawing_wcr.html
That said, having now thought a lot about graph layout and written a few papers on it my main takeaway is that whenever possible one shouldn't show graphs. They're always a mess for anything but the simplest structures (trees or near trees maybe). Instead, use graph structure to inform a different visualisation, or have an interactive graph explorer.
@2c47c3c1@56f80b5e Thanks! I think this course might have a bit more of a quantitative modelling and coding focus than you're looking for. I do think you're right to be sceptical about theories of the applicability of neuroscience to counselling because I don't think we know anywhere near enough about the brain to be able to make such recommendations with any degree of confidence. The course will feature a lot of me saying "we don't know how X works" and maybe that can help? It would be good if there was a course aimed at people in your situation, but I don't know of anything I'm afraid.
Oddbean new post about login | logout
Notes by Dan Goodman | export