Top Tweets for #doubledescent
At #SIAMAN25, @RishiSonthalia, @_akhilan, and @tj_jarvis delivered a series of presentations on #doubledescent that emphasized myriad ways to describe, analyze, and control the unique phenomenon. @maaminian recaps their talks in SIAM News. Check it out! https://t.co/osLhKkrQh5
El fenómeno #DoubleDescent en #MachineLearning revela que al aumentar datos o parámetros, el error puede inicialmente descender, luego subir y finalmente volver a descender, desafiando la intuición tradicional sobre el sobreajuste. @nicolas_araque
Google の論文「大規模言語モデルの能力創発 Emergent Abilities of Large Language Models」が話題になったので関連論文を幾つかまとめてみた
#OpenAI #ChatGPT #ScalingLaw #Pruning #LotteryTicketHypothesis #DoubleDescent #Grokking #PhaseTransition #MoreIsDifferent
https://t.co/iOTfVYGWpi
RT Double Descent and ML Models as Generalized Database Indexes https://t.co/U1VIAL6tkf #openai #intellectualproperty #deeplearning #opinion #doubledescent
📢 Overfitting + Underfitting = #DoubleDescent
Check out our #ICML2022 work on "Multi-scale Feature Learning Dynamics: Insights for Double Descent"
Thursday, 21 Jul 11:30 a.m. EDT, ICML 2022 Spotlight talks
w/ @AmartyaMitra, Yoshua Bengio, @g_lajoie_
https://t.co/XC2cSmCMwF
During the Advent season we also remember Jesus’ “double descent”–the divine descent to the manger, and a descent to the cross. You can’t separate one from the other. #doubledescent
A lottery winning read for the weekend:
Targeting a theory behind the #doubledescent. The paper by @jefrankle and Michael Carbin hypothesizes #winningticket sub-nets in every sufficiently deep #network which can perform just as well by itself. 🍀#AI #DNN https://t.co/j3PMgUf1ea
#DoubleDescent, the Lottery Ticket Hypothesis and "if is data is the new oil then #AI is the new CO2 - Key take aways from today's session with @johndkelleher at The International John McCarthy Summer School. #AISS @AdaptCentre @Microsoft
Nice historical discussion around the #doubledescent curve. Thanks Dmitry Kobak @hippopedoid. Next week, we'll publish a new blog post explaining how to reproduce the behavior of this curve in Python with LightOn's #OPU and #RandomFeatures.
Spent some time investigating history of "double descent". As a function of model complexity, I haven't seen it described before 2017. As a function of sample size, it can be traced to 1995; earlier research seems less relevant. Also: I think we need a better term. Thread. (1/n)
beautiful thread reconstructing the history of sightings of the two-step bias/variance trade-off, aka #doubledescent 👇🏽👇🏽👇🏽
Spent some time investigating history of "double descent". As a function of model complexity, I haven't seen it described before 2017. As a function of sample size, it can be traced to 1995; earlier research seems less relevant. Also: I think we need a better term. Thread. (1/n)
Last Seen Hashtags on Sotwe
Most Popular Users
Elon Musk 
@elonmusk
241.6M followers
Barack Obama
@barackobama
119M followers
Cristiano Ronaldo
@cristiano
114.1M followers
Donald J. Trump
@realdonaldtrump
111.8M followers
Narendra Modi
@narendramodi
107.2M followers
Rihanna
@rihanna
98.7M followers
NASA
@nasa
92.4M followers
Justin Bieber
@justinbieber
91.8M followers
KATY PERRY
@katyperry
89.8M followers
Taylor Swift
@taylorswift13
83.8M followers
Lady Gaga
@ladygaga
75.3M followers
Virat Kohli
@imvkohli
73M followers
Kim Kardashian
@kimkardashian
70.8M followers
YouTube
@youtube
68.8M followers
Neymar Jr
@neymarjr
66.1M followers
Bill Gates
@billgates
65M followers
Selena Gomez
@selenagomez
62.9M followers
The Ellen Show
@theellenshow
62.3M followers
CNN
@cnn
61.8M followers
X
@x
60.7M followers