Paul Todorov

Eng/Data/Product at Voodoo, former ML quant at JP Morgan

[Paul Todorov @PaulTodorov]

@tunguz @FrankRHutter @RichmanRonald @rasbt "xgboost" can mean many different things depending on feature engineering and tuning..
tabnet also shared this kind of benchmark in the past and it didn't last.
@tunguz do you plan to challenge these numbers by any chance ?

[Paul Todorov @PaulTodorov]

@bentossell @readwise save thread

[Paul Todorov @PaulTodorov]

@hardmaru @StabilityAI Congrats. Great match and great news for the ecosystem

[Paul Todorov @PaulTodorov]

@videodrome @artnet @briankboucher Btw the title of the article is 200% stupid since the fact that computer generated images can fool people has very little to do with AI art..Years have passed but most people commenting on this still have no clue about ML and fail to understand the true novelty of this approach

[Paul Todorov @PaulTodorov]

Super nouvelle pour nos amis à 4 pattes, avec le lancement de la campagne Ulule du projet Lopin !
https://t.co/1rkkWKYWTw

[Joseph Redmon @pjreddie]

I stopped doing CV research because I saw the impact my work was having. I loved the work but the military applications and privacy concerns eventually became impossible to ignore.

https://t.co/DMa6evaQZr

[Paul Todorov @PaulTodorov]

@artnome @ChristiesInc @DrBeef_ AI Art is in safe hands 👍🏻

[Artnome @artnome]

New Artnome Post: Why is AI Art Copyright So Complicated?

https://t.co/Q76pFh46Z0

[Jeremy Howard @jeremyphoward]

Wow wow wow! @GoogleColab now supports the fastai library directly :D That means you can run all the lessons at https://t.co/aQsW5afov6, and do your own research and development with fastai, all for free. That's... that's just amazing.

To get started: https://t.co/wop4GmJYSN

[Robbie Barrat @videodrome]

"There is little flesh here, only images of flesh"

[Paul Todorov @PaulTodorov]

@OriolVinyalsML I failed to reproduce the results from the « neural conversational model » https://t.co/8d20YzilQd ;)

[Paul Todorov @PaulTodorov]

Very cool python stuff from @rasbt :)

https://t.co/c5uHtQmQfY

[hardmaru @hardmaru]

“So you want to be a Research Scientist” by Vincent Vanhoucke @GoogleAI:
•Your will spend a career working on things that don’t work
•Your work will be obsolete the minute you publish it
•Your entire career will largely be measured by 1 number (H-Index)
https://t.co/rDROdZFxs3

[Sebastian Ruder @seb_ruder]

Great post about the most common supervised and unsupervised sentence representation methods.

[Paul Todorov @PaulTodorov]

Cool project by #Kraby and Olivier Large
#ML #Robotics
https://t.co/BCJXGcTdz7

[Paul Todorov @PaulTodorov]

@HSBC_UK It is still down. I cannot move my money. Have been waiting for 30 mins on the phone to talk to someone, in vain. Great service

[Paul Todorov @PaulTodorov]

@DrBeef_ 100% agree, your work has been stolen once again !😂

[Paul Todorov @PaulTodorov]

@glagolista @hardmaru Waiting for the blog post :)

[Sebastian Ruder @seb_ruder]

New NLP News: ML on code, Understanding RNNs, Deep Latent Variable Models, Writing Code for NLP Research, Quo vadis, NLP?, Democratizing AI, ML Cheatsheets, Spinning Up in Deep RL, Papers with Code, Unsupervised MT, Multilingual BERT https://t.co/6s2y4ABO2c via @revue

[Wieland Brendel @wielandbr]

Results of the NIPS Adversarial Vision Challenge are in! Common themes are low-frequency versions of the boundary attack, ensembling of defenses and new adversarial training schemes. Find all the details at https://t.co/LloXH3Y5v5