AI Weekly Update - February 7th, 2022

Thanks for watching! Please subscribe for more Deep Learning and AI videos, the list of papers is below under "Content Links" Please subscribe to SeMI Technologies (creators of the Weaviate Vector Search Engine) on YouTube: https://www.youtube.com/c/SeMI-and-Weaviate Content Links: Fully Online Meta-Learning without Task Boundaries: https://arxiv.org/abs/2202.00263 Datamodels: Predicting Predictions from Training Data: https://arxiv.org/abs/2202.00622 Adaptive Discrete Communication Bottlenecks with Dynamic Vector Quantization: https://arxiv.org/abs/2202.01334 Competition-Level Code Generation with AlphaCode: https://storage.googleapis.com/deepmind-media/AlphaCode/competition_level_code_generation_with_alphacode.pdf GPT-NeoX-20B: https://blog.eleuther.ai/announcing-20b/ PromptSource: https://arxiv.org/abs/2202.01279 Chain of Thought Prompting Elicits Reasoning in Large Language Models: https://arxiv.org/abs/2201.11903 Scaling Laws for Routed Language Models: https://arxiv.org/abs/2202.01169 Active Learning over Multiple Domains in Natural Language Tasks: https://arxiv.org/abs/2202.00254 Can Robots Follow Instructions for New Tasks? https://ai.googleblog.com/2022/02/can-robots-follow-instructions-for-new.html BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation: https://arxiv.org/pdf/2201.12086.pdf How to Leverage Unlabeled Data in Offline Reinforcement Learning: https://arxiv.org/abs/2202.01741 The Challenges of Exploration for Offline Reinforcement Learning: https://arxiv.org/pdf/2201.11861.pdf ETSformer: Exponential Smoothing Transformers for Time-series Forecasting: https://arxiv.org/pdf/2202.01381.pdf CoST: Contrastive Learning of Disentangled Seasonal-Trend Representations for Time Series Forecasting: https://arxiv.org/abs/2202.01575 Can Transformers be Strong Treatment Effect Estimators? https://arxiv.org/abs/2202.01336 Chapters: 0:00 Introduction 0:18 Weaviate Vector Search 0:34 Fully Online Meta-Learning