Search Results for author:
Found 6 papers, 1 papers with code
Ten Hard Problems in Artificial Intelligence We Must Get Right
We explore the AI2050 "hard problems" that block the promise of AI and cause AI risks: (1) developing general capabilities of the systems; (2) assuring the performance of AI systems and their training processes; (3) aligning system goals with human goals; (4) enabling great applications of AI in real life; (5) addressing economic disruptions; (6) ensuring the participation of all; (7) at the same time ensuring socially responsible deployment; (8) addressing any geopolitical disruptions that AI causes; (9) promoting sound governance of the technology; and (10) managing the philosophical disruptions for humans living in the age of AI.
Activation Addition: Steering Language Models Without Optimization
We demonstrate ActAdd on GPT-2 on OpenWebText and ConceptNet, and replicate the effect on Llama-13B and GPT-J-6B.
Massively Parallel Reweighted Wake-Sleep
Attaining such a large number of importance samples is intractable in all but the smallest models.
Decision trees compensate for model misspecification
These perform well in the presence of complex interactions, with tree depth governing the order of interactions.
Legally grounded fairness objectives
Recent work has identified a number of formally incompatible operational measures for the unfairness of a machine learning (ML) system.
How Robust are the Estimated Effects of Nonpharmaceutical Interventions against COVID-19?
To what extent are effectiveness estimates of nonpharmaceutical interventions (NPIs) against COVID-19 influenced by the assumptions our models make?
