all AI news
Unveiling Transformers with LEGO: a synthetic reasoning task. (arXiv:2206.04301v2 [cs.LG] UPDATED)
July 5, 2022, 1:11 a.m. | Yi Zhang, Arturs Backurs, Sébastien Bubeck, Ronen Eldan, Suriya Gunasekar, Tal Wagner
cs.LG updates on arXiv.org arxiv.org
We propose a synthetic task, LEGO (Learning Equality and Group Operations),
that encapsulates the problem of following a chain of reasoning, and we study
how the transformer architectures learn this task. We pay special attention to
data effects such as pretraining (on seemingly unrelated NLP tasks) and dataset
composition (e.g., differing chain length at training and test time), as well
as architectural variants such as weight-tied layers or adding convolutional
components. We study how the trained models eventually succeed at …
More from arxiv.org / cs.LG updates on arXiv.org
Regularization by Texts for Latent Diffusion Inverse Solvers
1 day, 10 hours ago |
arxiv.org
When can transformers reason with abstract symbols?
1 day, 10 hours ago |
arxiv.org
Jobs in AI, ML, Big Data
Data Scientist (m/f/x/d)
@ Symanto Research GmbH & Co. KG | Spain, Germany
Automated Greenhouse Expert - Phenotyping & Data Analysis (all genders)
@ Bayer | Frankfurt a.M., Hessen, DE
Machine Learning Scientist II
@ Expedia Group | India - Bengaluru
Data Engineer/Senior Data Engineer, Bioinformatics
@ Flagship Pioneering, Inc. | Cambridge, MA USA
Intern (AI lab)
@ UL Solutions | Dublin, Co. Dublin, Ireland
Senior Operations Research Analyst / Predictive Modeler
@ LinQuest | Colorado Springs, Colorado, United States