Share this page!

Enter URL or ID to Unroll

You can paste full URL like: https://x.com/threadreaderapp/status/1644127596119195649
or just the ID like: 1644127596119195649

How to get URL link on X (Twitter) App

  1. On the Twitter thread, click on or icon on the bottom
  2. Click again on or Share Via icon
  3. Click on Copy Link to Tweet
  4. Paste it above and click "Unroll Thread"!
  5. More info at Twitter Help
Christopher Rozell Profile picture
@crozSciTech
Sep 7, 2018 • 3 tweets • 2 min read • Read on X
"Nothing in computational neuroscience makes since except in light of optimization" @dyamins #preach #ccn18
Also, "I don't care about neurons." @gallantlab #ccn18
errr....."since" = "sense".

• • •

Missing some Tweet in this thread? You can try to force a refresh
　

Keep Current with Christopher Rozell

Christopher Rozell Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @crozSciTech

Christopher Rozell Profile picture
Sep 24, 2020
New neuro theory manuscript up on bioRxiv titled "Constrained brain volume in an efficient coding model explains the fraction of excitatory and inhibitory neurons in sensory cortices". With @INemenman and led by Arish Alreja. Take a look and RT!

biorxiv.org/content/10.110…
TL;DR Mammalian brains vary in size by orders of magnitude. But the ratio of excitatory to inhibitory cells in their sensory cortexes varies by only a factor of 3. In a new preprint, we explain why this could be. Read below to find out.
To help us understand circuit computations, we looked at structural properties conserved across species. For example, while the number of neurons in primary sensory areas varies by 100x, the ratio of E:I neurons varies by less than 3x.
Read 19 tweets
Christopher Rozell Profile picture
Sep 22, 2020
New paper just accepted at IEEE Transactions in Computational Imaging titled "Parallel Unbalanced Optimal Transport Regularization for Large-scale Imaging Problems" with @John07252731 and Nick Bertrand (not on twitter): arxiv.org/pdf/1909.00149…
Code is available here: github.com/siplab-gt/UOT-…
We have used Optimal Transport (aka Earth Mover's Distance) as a regularizer for inverse problems and sparse tracking problems in the past (arxiv.org/abs/1806.04674).
Read 8 tweets
Christopher Rozell Profile picture
May 30, 2019
ML (and maybe neuro) folks, we're excited about a new ICML paper: "Active Embedding Search via Noisy Paired Comparisons" by Canal (@GregHCanal), Massimino, Davenport and Rozell. What's the summary? Glad you asked.

Preprint: arxiv.org/pdf/1905.04363
Code: github.com/siplab-gt/pair…
The problem we're considering is "preference search", which is the task of finding the best "item" for a given user. An "item" is just a vector in a vector space, so this is a very general formulation that could have many applications.
Items could be material things (e.g., a product, an image in a dataset), or it could be something abstract like the preferred settings of a medical device or model parameters. Especially interested in suggestions of neuro applications.
Read 11 tweets
Christopher Rozell Profile picture
Dec 2, 2018
@resolvingdust @sbarolo I agree that the attention need to be turned to those women. My question is, for a man who is (wrongly) focused on the potential for false accusations, what ways are there to get him focused on the right things (the victims)?
@resolvingdust @sbarolo These guys are real people that are causing real damage in their organizations by this attitude. Getting them to leave the twitter conversation doesn't change the damage they are doing there.
@resolvingdust @sbarolo I don't know what will change them, but I would like to try talking to them. I want them to hear about @sbarolo's extensive list. But they have to be listening in order to get them to actually hear about these thousands of women and essentially no false accusations.
Read 4 tweets
Christopher Rozell Profile picture
Dec 2, 2018
@resolvingdust @j_zelikova @sbarolo @John7Istheman @McLNeuro @james_t_webber So my shorthand to try and communicate the idea on twitter is obviously conveying something I didn't intend. I'm sorry. I was trying to say we share similar goals.
@resolvingdust @j_zelikova @sbarolo @John7Istheman @McLNeuro @james_t_webber My hope is that women would face fewer instances of harassment in the institutions, organizations and communities where they engage in STEM careers.
@resolvingdust @j_zelikova @sbarolo @John7Istheman @McLNeuro @james_t_webber The role I'm trying to play in that is to talk to other men who don't appear aware of the magnitude of the issue and try to get them to a point where they can recognize that.
Read 6 tweets
Christopher Rozell Profile picture
Mar 20, 2018
We often want to understand a complex dynamical system (e.g., climate) by collecting measurements over time of a relevant quantity /1
(e.g., carbon dioxide concentration). Previous (amazing) results showed that this time series can be used to create an image of the /2
hidden system, but it didn't tell us if that image was good in the sense that it faithfully preserved the geometry of the system's /3
Read 5 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

Send Email!

:(