Top Tweets for #RLDM2019
Saw @fierycushman give a talk on this at #RLDM2019, and found it hugely insightful! Such an interesting line of research!
There is literally no limit to the thoughts the human mind can have. But then out of that infinity of possible thoughts, how do we know which ones to consider and which ones aren't worth our time? How do we know what *not* to think? Here's our new idea: https://t.co/iqno0ahsli
@phillipsjs Time pressure is also a huge issue in the AI ethics domains we're interested in, about which I've mused in this blog post:
https://t.co/lqz7Q9AzX8
I found @fierycushman's talk on your work at #RLDM2019 to be super insightful, and have been chatting this work up to colleagues.
After some delays--through an unplanned but productive meetings with @DanielaJPalombo--am now back in the UK!!
Lots of exciting ideas after #RLDM2019 and visiting people in Edmonton, time to get some projects written up.
One of my favourite talks from #RLDM2019! These ideas are essential to RL in any non game/simulated robotics scenarios, and opens up a huge space for research. Thank you :)
Our ability to imagine the unknown is a critical piece of our creativity & intelligence. Counterfactual /batch RL seeks to replicate this & has huge potential benefits. It's great to be in this growing area & I share a few ideas in my tutorial: https://t.co/VOmrRE4gv3
#RLDM2019
RLDM is just lovely! It was great to see everyone and hear about all the wonderful RL work going on -- already looking forward to RLDM 2021.
Notes from this round available here: https://t.co/wWrNRc9hjC #RLDM2019
Thank you our amazing speakers and participants for the Best workshop ever (yes, Iβm biased) at #RLDM2019 Sign up here if you would like to receive a summary of the workshop and discussions https://t.co/fKEEMCRFtv
Reinforcement Learning and Decision Making #RLDM2019
Montreal, Canada
Notes by David Abel: https://t.co/fEWnFNUUX2
#ReinforcementLearning #DecisionMaking #MachineLearning
#ai
I was super happy to learn this week that the extended abstract version of this work got a Best Paper Award at #RLDM2019! Congrats to my colleagues @LiamFedus, @carlesgelada, Yoshua Bengio and @marcgbellemare for making this work possible!
Fresh on arXiv: Hyperbolic Discounting and Learning Over Multiple Horizons
We question the RL paradigm of discounting by a single discount factor, gamma. Modeling many Q-values allows you to hyperbolically-discount and is also a great auxiliary task.
https://t.co/3XmXVwB1qW
Thanks to all who contributed to making this workshop at #RLDM2019 a great success! Thanks especially to the conference organizers including workshop chairs Philip Thomas and @gershbrain, and to our invited speakers @mollycrockett and @alex_peys!
#RLDM2019 was sooooo much fun hanging with old friends, making some new ones, and seeing where the field is going. But nothing like coming home to these guys at the end of the day πππ
V(RLDM) <-- V(RLDM) + Ξ±[warm_fuzzy_feelings + Ξ³(post_conference_high - pre_conference_panic)]
#RLDM2019 was exactly the reminder I needed that studying the mind (all minds!) is the coolest thing & we're so lucky to do it. Is it 2021 yet? π§ π€π
The former advisor and advisee have some healthy disagreements on the necessity of the concept of intrinsic motivation.
I think Rich believes that intrinsic motivation is a misnomer; reward is all what we have and need.
#RLDM2019
RLDM is just lovely! It was great to see everyone and hear about all the wonderful RL work going on -- already looking forward to RLDM 2021.
Notes from this round available here: https://t.co/wWrNRc9hjC #RLDM2019
Satinder Singh from @michigan_AI talks about how to optimize an intrinsic reward function at #RLDM2019.
Or: let's evolve the internal reward function of an agent in order to increase its chance of survival (objective reward).
Doina Precup from @rllabmcgill @MILAMontreal @mcgillu talks about new directions in exploration in reinforcement learning.
#RLDM2019 workshop on Curiosity in Decision Making.
#RLDM2019 was brilliant. Tons of great talks, posters and people. Thanks to those of you that gave feedback on our poster (preprint here https://t.co/rsAflImgDz). And thanks to @Royalcom1851 and @dunnhumby for allowing this all to happen :)
Thatβs a wrap! Thanks for coming to visit us at #RLDM2019 in #Montreal, and see you soon at @ACL2019_Italy. To learn more about our open job opportunities, visit our careers page: https://t.co/8OCV410tlA
Thanks for getting excited about brains with me yesterday, friends!! If you have #RLDM2019 FOMO and/or missed the poster session, a PDF is available here: https://t.co/naFpdldEXR
Really enjoyed Fiery Cushman's talk at #RLDM2019 . This notion of generating a set of candidate choices or actions more habitually, and then deliberating on them seems very intuitive, and also makes a lot of sense of a computational approach
Last Seen Hashtags on Sotwe
snapchat
Seen from Thailand
ozgetore
Seen from United States
minichat((((((()))))))*
Seen from Turkey
motherdaughter incest
Seen from Turkey
minichat
Seen from Turkey
poopinggirls
Seen from Turkey
fe2fanart
Seen from Australia
ipcam
Seen from United States
adanatravesti travesti
Seen from Turkey
sm46031666
Seen from United States
Most Popular Users
Elon Musk 
@elonmusk
241.7M followers
Barack Obama
@barackobama
119M followers
Cristiano Ronaldo
@cristiano
114.5M followers
Donald J. Trump
@realdonaldtrump
111.9M 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
90.1M followers
Taylor Swift
@taylorswift13
84M followers
Lady Gaga
@ladygaga
75.5M followers
Virat Kohli
@imvkohli
73.5M followers
Kim Kardashian
@kimkardashian
70.9M followers
YouTube
@youtube
68.8M followers
Neymar Jr
@neymarjr
66.5M followers
Bill Gates
@billgates
65.2M followers
Selena Gomez
@selenagomez
63.1M followers
The Ellen Show
@theellenshow
62.3M followers
CNN
@cnn
61.8M followers
X
@x
60.7M followers