Active Optimization and Self Driving Cars
نویسنده
چکیده
Self driving cars hold the promise of making transportation safer and more efficient than ever before possible. They will also be among the most complex robotic systems ever fielded and thus require an unprecedented level of machine learning throughout to achieve their desired performance. These learners create an everchanging environment for all algorithms operating in the system and optimizing their performance will become a perpetual activity rather than a one-off task. I will present active optimization methods and their application in robotics applications, focusing on scaling up the dimensionality and managing multi-fidelity evaluations. I will summarize lessons learned and thoughts on future directions as these methods move into fielded systems. Finally, I will describe current efforts on self driving cars and give some views on the future of them. BIOGRAPHY Dr. Jeff Schneider is the engineering lead for machine learning at Uber's Advanced Technologies Center. He is currently on leave from Carnegie Mellon University where he is a research professor in the School of Computer Science. He has 20 years experience developing, publishing, and applying machine learning algorithms in government, science, and industry. He has over 100 publications and regularly gives talks and tutorials on the subject. Previously, Jeff was the co-founder and CEO of Schenley Park Research, a company dedicated to bringing machine learning to industry. Later, he developed a machine learning based CNS drug discovery system and commercialized it during two years as Psychogenics' Chief Informatics Officer. Through his research, commercial, and consulting efforts, he has worked with dozens of companies and government agencies around the world. Appears in: Proc. of the 16th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2017), S. Das, E. Durfee, K. Larson, M. Winikoff (eds.), May 8–12, 2017, São Paulo, Brazil. Copyright © 2017, International Foundation for Autonomous Agents and Multiagent Systems (www.ifaamas.org). All rights reserved.
منابع مشابه
Disruption of Individual Mobility Ahead? A Longitudinal Study of Risk and Benefit Perceptions of Self-Driving Cars on Twitter
In this paper, we address the question if there is a disruption of individual mobility by self-driving cars ahead. In order to answer this question, we take the user perspective and conduct a longitudinal study of social media data about self-driving cars from Twitter. The study analyzes 601,778 tweets from March 2015 to July 2016. We use supervised machine learning classification to extract re...
متن کاملSelf-Driving Cars - An AI-Robotics Challenge
In recent years, all major automotive companies have launched initiatives towards cars that assist people in making driving decisions. The ultimate goal of all these efforts are cars that can drive themselves. The benefit of such a technology could be enormous. At present, some 42,000 people die every year in traffic accidents in the U.S., mostly because of human error. Self-driving cars could ...
متن کاملA Philosophy for Developing Trust in Self‐ Driving Cars
For decades, our lives have depended on the safe operation of automated mechanisms around and inside us. The autonomy and complexity of these mechanisms is increasing dramatically. Autonomous systems such as self-driving cars rely heavily on inductive inference and complex software, both of which confound traditional software-safety techniques that are focused on amassing sufficient confirmator...
متن کاملThe Effect of System Restrictions on Acceptance of Self-Driving Cars
Only some state of the art driving automation systems (DAS) of current self-driving cars include system restrictions that actively counteract misuse instead of only informing the drivers of the limitations of the system. Misuse can lead to accidents, which could reduce acceptance of self-driving cars at a population level. However, system restrictions could also affect acceptance as DAS with re...
متن کاملResponsibility and the Moral Phenomenology of Using Self-Driving Cars
This paper explores how the phenomenology of using selfdriving cars influences conditions for exercising and ascribing responsibility. First, a working account of responsibility is presented, which identifies two classic Aristotelian conditions for responsibility and adds a relational one, and which makes a distinction between responsibility for (what one does) and responsibility to (others). T...
متن کامل