Published papers at TMLR · @tmlrpub
564 followers · 593 posts · Server sigmoid.social
Representations and Computations in Transformers that Support Generalization on Structured Tasks
Yuxuan Li, James McClelland
Action editor: Stefan Lee.
#attention #learns #representations
New Submissions to TMLR · @tmlrsub
205 followers · 745 posts · Server sigmoid.social
When Less is More: Simplifying Inputs Aids Neural Network Understanding
#simplicity #distractors #learns
Published papers at TMLR · @tmlrpub
563 followers · 566 posts · Server sigmoid.social
FairGrad: Fairness Aware Gradient Descent
Gaurav Maheshwari, Michaël Perrot
Action editor: Novi Quadrianto.
New Submissions to TMLR · @tmlrsub
199 followers · 699 posts · Server sigmoid.social
SkillS: Adaptive Skill Sequencing for Efficient Temporally-Extended Exploration
#learns #reinforcement #exploration
New Submissions to TMLR · @tmlrsub
190 followers · 648 posts · Server sigmoid.social
Simplifying and Understanding State Space Models with Diagonal Linear RNNs
New Submissions to TMLR · @tmlrsub
185 followers · 641 posts · Server sigmoid.social
Two-Level Actor-Critic Using Multiple Teachers
#learns #reinforcement #teachers
beSpacific · @bespacific
907 followers · 1746 posts · Server newsie.social#AI Is a Lot of Work. As the technology becomes #ubiquitous, a vast #tasker #underclass is emerging and not going anywhere. #AI #learns by finding #patterns in enormous quantities of #data, but first that data has to be sorted, #tagged by people, a vast #workforce mostly hidden behind machines. It’s difficult and #repetitive work. A several-second blip of footage took eight hours to annotate, for which a college graduate in #Nairobi was paid about $10. https://www.theverge.com/features/23764584/ai-artificial-intelligence-data-notation-labor-scale-surge-remotasks-openai-chatbots?tpcc=NL_Marketing
#ai #ubiquitous #tasker #underclass #learns #patterns #data #tagged #workforce #repetitive #nairobi
Published papers at TMLR · @tmlrpub
520 followers · 448 posts · Server sigmoid.social
Reinforcement Teaching
Calarina Muslimani, Alex Lewandowski, Dale Schuurmans, Matthew E. Taylor, Jun Luo
Action editor: Marcello Restelli.
#learns #reinforcement #learnable
New Submissions to TMLR · @tmlrsub
178 followers · 595 posts · Server sigmoid.social
DCP: Learning Accelerator Dataflow for Neural Network via Propagation
New Submissions to TMLR · @tmlrsub
180 followers · 563 posts · Server sigmoid.social
GraphPNAS: Learning Probabilistic Graph Generators for Neural Architecture Search
Published papers at TMLR · @tmlrpub
517 followers · 410 posts · Server sigmoid.social
Lightweight Learner for Shared Knowledge Lifelong Learning
Yunhao Ge, Yuecheng Li, Di Wu et al.
Action editor: Edward Grefenstette.
Published papers at TMLR · @tmlrpub
516 followers · 377 posts · Server sigmoid.social
Do Vision-Language Pretrained Models Learn Composable Primitive Concepts?
Tian Yun, Usha Bhalla, Ellie Pavlick, Chen Sun
Action editor: Zhe Gan.
#concepts #learns #Recognition
New Submissions to TMLR · @tmlrsub
172 followers · 513 posts · Server sigmoid.social
FL Games: A federated learning framework for distribution shifts
New Submissions to TMLR · @tmlrsub
170 followers · 497 posts · Server sigmoid.social
Lightweight Learner for Shared Knowledge Lifelong Learning
New Submissions to TMLR · @tmlrsub
167 followers · 470 posts · Server sigmoid.social
Published papers at TMLR · @tmlrpub
507 followers · 309 posts · Server sigmoid.social
L-SVRG and L-Katyusha with Adaptive Sampling
Boxin Zhao, Boxiang Lyu, mladen kolar
New Submissions to TMLR · @tmlrsub
160 followers · 367 posts · Server sigmoid.social
SLM: End-to-end Feature Selection via Sparse Learnable Masks
Published papers at TMLR · @tmlrpub
498 followers · 234 posts · Server sigmoid.social
Recurrent networks, hidden states and beliefs in partially observable environments
Gaspard Lambrechts, Adrien Bolland, Damien Ernst
#reinforcement #learns #recurrent
New Submissions to TMLR · @tmlrsub
159 followers · 343 posts · Server sigmoid.social
Do Vision-Language Pretrained Models Learn Composable Primitive Concepts?
#concepts #learns #Recognition
Published papers at TMLR · @tmlrpub
498 followers · 223 posts · Server sigmoid.social
No More Pesky Hyperparameters: Offline Hyperparameter Tuning for RL
Han Wang, Archit Sakhadeo, Adam M White et al.
#hyperparameters #hyperparameter #learns