Goto

Collaborating Authors

ijcai-pricai 2020


One Hundred Year Study on Artificial Intelligence (AI100) – a panel discussion at #IJCAI-PRICAI 2020

AIHub

The mission of AI100 is to launch a study every five years, over the course of a century, to better track and anticipate how artificial intelligence propagates through society, and how it shapes different aspects of our lives. This IJCAI session brought together some of the people involved in the AI100 initiative to discuss their efforts and the direction of the project. The goals of the AI100 are "to support a longitudinal study of AI advances on people and society, centering on periodic studies of developments, trends, futures, and potential disruptions associated with the developments in machine intelligence, and formulating assessments, recommendations and guidance on proactive efforts". Working on the AI100 project are a standing committee and a study panel. The first study panel report, released in 2016, can be read in full here.


Choice Set Misspecification in Reward Inference

arXiv.org Artificial Intelligence

Specifying reward functions for robots that operate in environments without a natural reward signal can be challenging, and incorrectly specified rewards can incentivise degenerate or dangerous behavior. A promising alternative to manually specifying reward functions is to enable robots to infer them from human feedback, like demonstrations or corrections. To interpret this feedback, robots treat as approximately optimal a choice the person makes from a choice set, like the set of possible trajectories they could have demonstrated or possible corrections they could have made. In this work, we introduce the idea that the choice set itself might be difficult to specify, and analyze choice set misspecification: what happens as the robot makes incorrect assumptions about the set of choices from which the human selects their feedback. We propose a classification of different kinds of choice set misspecification, and show that these different classes lead to meaningful differences in the inferred reward and resulting performance. While we would normally expect misspecification to hurt, we find that certain kinds of misspecification are neither helpful nor harmful (in expectation). However, in other situations, misspecification can be extremely harmful, leading the robot to believe the opposite of what it should believe. We hope our results will allow for better prediction and response to the effects of misspecification in real-world reward inference.


Tweet round-up from #IJCAI-PRICAI 2020

AIHub

The 29th International Joint Conference on Artificial Intelligence and the 17th Pacific Rim International Conference on Artificial Intelligence (IJCAI-PRICAI 2020) is in full swing. The event started on 7 January and will run until 15 January. Here, we capture the first few days of the conference through tweets from attendees. Wandering around the #IJCAI2020 venue is fun! It's basically equipped with most of what's in the physical venue -- coffee, pub, beach (when you are tired), and the gate to hyperspace! pic.twitter.com/dIzwn2MZmu


What's coming up at IJCAI-PRICAI 2020?

AIHub

IJCAI-PRICAI2020, the 29th International Joint Conference on Artificial Intelligence and the 17th Pacific Rim International Conference on Artificial Intelligence starts today and will run until 15 January. Find out what's happening during the event. The conference schedule is here and includes tutorials, workshops, invited talks and technical sessions. There are also competitions, early career spotlight talks, panel discussions and social events. There will be eight invited talks on a wide variety of topics.