Name: Approximately Correct Machine Intelligence (ACMI) Lab
Type: Organization
Bio: Research on machine learning, its social impacts, and applications to healthcare. PI—@zackchase
Twitter: acmi_lab
Blog: http://acmilab.org/
Approximately Correct Machine Intelligence (ACMI) Lab's Projects
Code and results accompanying our paper titled CHiLS: Zero-Shot Image Classification with Hierarchical Label Sets
Official repository for CMU Machine Learning Department's 10717: "The Art of the Paper".
Official repository for CMU Machine Learning Department's 10721: "Philosophical Foundations of Machine Intelligence".
Official repository for CMU Machine Learning Department's 10732: Robustness and Adaptivity in Shifting Environments
Learning the Difference that Makes a Difference with Counterfactually-Augmented Data
Python package for Evaluating Medical Datasets Over Time (EMDOT).
Official implementation for the ICLR 2023 paper: Disentangling the Mechanisms Behind Implicit Regularization in SGD.
Code and results accompanying our paper titled Unsupervised Learning under Latent Label Shift at NeurIPS 2022
Code for the paper "Local Causal Discovery for Estimating Causal Effects".
Code accompanying our paper "Domain Adaptation under Missingness Shift" at AISTATS 2023.
code for modular summarization work published in ACL2021 by Krishna et al
Official Repository for the paper "Model-tuning Via Prompts Makes NLP Models More Robust"
Code for the paper "Efficient Online Estimation of Causal Effects by Deciding What to Observe".
Code accompanying our paper titled Online Label Shift: Optimal Dynamic Regret meets Practical Algorithms
Code and results accompanying our paper titled Domain Adaptation under Open Set Label Shift
A Curated Set of Papers Actually Worth Reading
Pretraining summarization models using a corpus of nonsense
Code and results accompanying our paper titled Mixture Proportion Estimation and PU Learning: A Modern Approach at Neurips 2021 (Spotlight)
Code and results accompanying our paper titled RATT: Leveraging Unlabeled Data to guarantee generalization at ICML 2021 (Long Talk)
Code and results accompanying our paper titled RLSbench: Domain Adaptation under Relaxed Label Shift
Thrilling tales of heroic feats by ML's larger-than-life champions.