CS 188 notes IV

CS 188 notes IV

Inference with Bayes’ Rule

Models describe how (a portion of) the world works

"All models are wrong; but some are useful.” – George E. P. Box

SLAM: Simultaneous Localization And Mapping

DP-SLAM Austin Eliazar and Ronald Parr https://users.cs.duke.edu/~parr/dpslam/

Evaluation and observability for AI systems https://www.evidentlyai.com/

Very loose inspiration: human neurons

Show, Attend and Tell: Neural Image Caption Generation with Visual Attention https://arxiv.org/abs/1502.03044

Long-term Recurrent Convolutional Networks for Visual Recognition and Description https://arxiv.org/abs/1411.4389

Deep Visual-Semantic Alignments for Generating Image Descriptions https://arxiv.org/abs/1412.2306

Object Detection https://docs.ultralytics.com/tasks/detect/#predict

Efficient Estimation of Word Representations in Vector Space https://arxiv.org/abs/1301.3781

The Belmont ReportOffice of the SecretaryEthical Principles and Guidelines for the Protection of Human Subjects of ResearchThe National Commission for the Protection of Human Subjects of Biomedical and Behavioral ResearchApril 18, 1979 https://www.hhs.gov/ohrp/regulations-and-policy/belmont-report/read-the-belmont-report/index.html

Virtual Personas for Language Models via an Anthology of Backstories https://arxiv.org/abs/2407.06576

The leap in general-purpose capabilities from language models is a landmark in the development of Artificial Intelligence. The nature of this technology has led to a different set of research questions than are the typical focus of related venues. COLM is an academic venue focused on the study of language modeling, broadly defined, with the goal of creating a community of researchers with expertise in different disciplines, focused on understanding, improving, and critiquing the development of LM technology. https://colmweb.org/

Cross-Domain Imitation Learning via Optimal Transport https://arnaudfickinger.github.io/gwil/

Oasis: A Universe in a Transformer https://oasis-model.github.io/

Genie 2: A large-scale foundation world model https://deepmind.google/blog/genie-2-a-large-scale-foundation-world-model/

Responsible AI Initiative https://re-ai.berkeley.edu/home

Tokenizer https://platform.openai.com/tokenizer

Generative AI exists because of the transformer https://ig.ft.com/generative-ai/