This was a hugely collaborative effort, led by Meire Fortunato, Ryan Faulkner, Melissa Tan, and myself under @BlundellCharles's leadership, with Adrià Badia, Gavin Buttimore & @bigblueboo also playing key roles. Many more of my @DeepMindAI colleagues were also very supportive 2/n
-
-
Prikaži ovu nit
-
Results! 1) Some of these tasks are hard! Underfitting is still an issue in RL 2) Extrapolation isn't impossible for Deep RL agents, but it requires the right inductive biases and is far from solved 3) Adding a contrastive loss to an external memory is a good thing to do 3/npic.twitter.com/LEbaXFtcqG
Prikaži ovu nit -
Tasks! In addition to a standard train/test split based on partitioning some variable (e.g. color), we also pick a scalar variable (e.g. size of room). We can thus train on some values and test on unseen values inside the range (interp) or outside of the range (extrap) 4/npic.twitter.com/RDKaVbQWlz
Prikaži ovu nit -
Memory Recall Agent! A new agent that combines 1) an external memory 2) contrastive auxiliary loss 3) jumpy-backpropagation for credit assignment Importantly, all of these pieces were validated through over 10 ablations! 5/npic.twitter.com/5rQjDjQVYA
Prikaži ovu nit -
Feverishly working on preparing the tasks for an external just in time for
@NeurIPSConf. We hope these tasks represent an interesting challenge for the deep RL community. Excited to see what y'all can do with them! http://sites.google.com/corp/view/memory-tasks-suite/ … n/n back to work timepic.twitter.com/eurQnUVA2RPrikaži ovu nit
Kraj razgovora
Novi razgovor -
-
-
This deep dive on memory and generalisation is a really important direction for moving RL forward. Nice work!
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.