Things you're allowed to do. For reference. Ping me if I've forgotten any.
Follow
Milan Cvitkovic (visiting SF soon)
@MWCvitkovic
Neurotechnology | Focused Research Organizations
Milan Cvitkovic (visiting SF soon)’s Tweets
I'm probably late to the game, but I didn't realize ChatGPT knew SMILES
2
25
Congratulations to Jeff Diberto, Ph.D.! Jeff successfully defended his thesis, “Discovery Campaigns for Pharmacologically Unique Agonists at the Mu-Opioid and 5-Hydroxytryptamine 2A Receptors,” on Thursday, January 19th, under the direction of Professor Bryan Roth
9
40
Visiting SF soon! Whom must I simply meet?
6
1
16
Are there examples of neuropsych treatments systematically causing people to lie on questionnaires?
Especially lying in the positive direction. So it looked like the Tx was helping, but actually it was making subjects terrified of disappointing the researcher or something.
1
7
Would love to hear other people's ideas in this 🧵 too!
3
3
Show this thread
8. Inject false or impossible cards randomly into your deck. Keep your epistemics up! If you know there's always an answer, you'll get complacent!
2
1
19
Show this thread
7. This, but phrase the flashcard in the style of a famous person. I want Feynman to grill me on physics in a Feynman way
Quote Tweet
2. Change the phrasing of a card each time I see it (but with exact same information). "Muscle confusion" for the brain.
Show this thread
1
9
Show this thread
If anyone wants to collaborate on some of these lmk btw
3
1
11
Show this thread
6. Automatically merge related cards into harder mega-cards to force chunking. Also maybe automatically split cards into smaller concepts if I keep blowing them.
2
1
10
Show this thread
5. First, enable me to give verbal answers via transcription. Then, keep a record of my past answers, and force me to give a new answer/phrasing to a card if my past answers are too similar.
1
5
Show this thread
4.
Quote Tweet
we need step 5: have GPT-3 generate nicely formatted Anki cards you can choose to retain or not. twitter.com/bakztfuture/st…
3
10
Show this thread
3. Enable interactive cards, like short convos in a new language or multi-step terminal commands. Okay sure maybe this is beyond Anki, but still.
1
3
Show this thread
2. Change the phrasing of a card each time I see it (but with exact same information). "Muscle confusion" for the brain.
2
1
42
Show this thread
1. Obviously: generate from articles or my notes.
People are playing with this, but I bet many best practices to learn
1
1
15
Show this thread
Are there any good Anki + LLM integrations yet?
There are so many features I want: 🧵1/n
18
11
179
Show this thread
One year ago we had just opened our lab. Today we share our latest images of axons through a slice of mouse brain. Each color identifies a unique neuron. #FluorescenceFriday
11
53
There are some open positions at CR and FROs, please check them out!
3
8
A lot of modern neurotech projects are rooted in half a century of research.
DARPA funded a Harvard team in 1976 to build a typewriter that could be controlled with slight finger twitches + a Stanford team to infer words from brain activity.
2
7
30
Show this thread
How did art and art criticism become two separate jobs? And why doesn't science have this division of labor?
2
20
This goes into a list of best paper titles
4
105
878
Show this thread
Seems like there's a massive arb taking AI interpretability results/techniques and applying them to neuroscience data.
Anyone know of work on this?
3
2
15
Replying to
Anecdotal, but I've stopped getting colds/flu from travel since starting to use these immediately post-flight (and mid-flight for long flights), plus lots of hand washing.
a.co/d/192GlAJ
a.co/d/h54YxIg
2
1
Visiting Miami soon. Who should I meet?
4
4
Are there journals that have cutoffs on effect size rather than p-value?
2
8
Bounty: I'll send you $2 for every number (with citation) you send me that I can add to this list
4
2
10
Show this thread
Also this conversation suffers from a lack of clear definitions.
BCI, WBE, uploading, AIs, humans, human values: people are throwing around these terms like we have any operational definitions for them.
It's getting almost as bad as a free-will debate...
1
1
3
Anyone I should meet in London or Oxford next week?
11
1
18
we're excited to take the covers off our first product program, an optogenetic retinal prosthesis!
42
159
807
Show this thread
Who should I meet at SfN this weekend?
Super excited to share a preprint! At , we build the first closed-loop brain-machine (#BMI) interface that is able to decode internal speech 💭. Check it out here: medrxiv.org/content/10.110
32
392
1,472
Show this thread
Quote Tweet
I think the traditional tenure track in universities is roughly like the management track in big companies. There should be something like the IC track for universities. Research scientists who have the prestige of tenure track but without the administrative burden
Show this thread
1
If it turns out this was all a mistake caused by a grad student testing their drugs on the lab GC-MS, BUT that ketamine still works as an anthelmintic, I will be so happy.
3
Show this thread
It's not a bioprospecting party until someone finds a dissociative
1
5
Show this thread
Scientists in NY, FYI: a NY Public Library card gives you online access to Nature journals. Handy for when SciHub doesn't have the article yet.
ht for the tip!
1
1
17
Like, formally verified secured VMs, Butlerian regulation, some neurotechnology, etc.
3
8
Show this thread