Artificial intelligence is entering our lives in many ways on our smartphones, in our homes, in our cars. These systems can help people make appointments, drive and even diagnose illnesses. But as it continues to serve important and collaborative roles in peoples lives, a natural question is: Can I trust them? How do I know they will do what I expect?
Explainable artificial intelligence is a branch of artificial intelligence research that examines how artificial agents can be made more transparent and trustworthy to their human users. Trustworthiness is essential if robots and people are to work together. It seeks to develop systems that human beings find trustworthy while also performing well to fulfill designed tasks.
At the Center for Vision, Cognition, Learning, and Autonomy at University of California Los Angeles, we and our colleagues are interested in what factors make machines more trustworthy, and how well different learning algorithms enable trust. Our lab uses a type of knowledge representation a model of the world that artificial intelligence uses to interpret its surroundings and make decisions that can be more easily understood by humans. This naturally aids in explanation and transparency, thereby improving trust of human users.
In our latest research, we experimented with different ways a robot could explain its actions to a human observer. Interestingly, the forms of explanation that fostered the most human trust did not correspond to the learning algorithms that produced the best task performance. This suggests performance and explanation are not inherently dependent upon each other optimising for one alone may not lead to the best outcome for the other. This divergence calls for robot designs that takes into account both good task performance and trustworthy explanations.
In undertaking this study, our group was interested in two things. How does a robot best learn to perform a particular task? Then, how do people respond to the robots explanation of its actions?
We taught a robot to learn from human demonstrations how to open a medicine bottle with a safety lock. A person wore a tactile glove that recorded the poses and forces of the human hand as it opened the bottle. That information helped the robot learn what the human did in two ways: symbolic and haptic. Symbolic refers to meaningful representations of your actions: for example, the word grasp. Haptic refers to the feelings associated with your bodys postures and motions: for example, the sensation of your fingers closing together.
First, the robot learned a symbolic model that encodes the sequence of steps needed to complete the task of opening the bottle. Second, the robot learned a haptic model that allows the robot to imagine itself in the role of the human demonstrator and predict what action a person would take when encountering particular poses and forces.
It turns out the robot was able to achieve its best performance when combining the symbolic and haptic components. The robot did better using knowledge of the steps for performing the task and real-time sensing from its gripper than using either alone.
Now that the robot knows what to do, how can it explain its behavior to a person? And how well does that explanation foster human trust?
To explain its actions, the robot can draw on its internal decision process as well as its behavior. The symbolic model provides step-by-step descriptions of the robots actions, and the haptic model provides a sense of what the robot gripper is feeling.
In our experiment, we added an additional explanation for humans: a text write-up that provided a summary after the robot has finished attempting to open the medicine bottle. We wanted to see if summary descriptions would be as effective as the step-by-step symbolic explanation to gain human trust.
We asked 150 human participants, divided into four groups, to observe the robot attempting to open the medicine bottle. The robot then gave each group a different explanation of the task: symbolic, step-by-step, haptic arm positions and motions, text summary, or symbolic and haptic together. A baseline group observed only a video of the robot attempting to open the bottle, without providing any additional explanations.
We found that providing both the symbolic and haptic explanations fostered the most trust, with the symbolic component contributing the most. Interestingly, the explanation in the form of a text summary didnt foster more trust than simply watching the robot perform the task, indicating that humans prefer robots to give step-by-step explanations of what theyre doing.
The most interesting outcome of this research is that what makes robots perform well is not the same as what makes people see them as trustworthy. The robot needed both the symbolic and haptic components to do the best job. But it was the symbolic explanation that made people trust the robot most.
This divergence highlights important goals for future artificial intelligence and robotics research: to focus on pursuing both task performance and explainability. Only focussing on task performance may not lead to a robot that explains itself well. Our lab uses a hybrid model to provide both high performance and trustworthy explanations.
Performance and explanation do not naturally complement each other, so both goals need to be a priority from the start when building artificial intelligence systems. This work represents an important step in systematically studying how human-machine relationships develop, but much more needs to be done. A challenging step for future research will be to move from I trust the robot to do X to I trust the robot.
For robots to earn a place in peoples daily lives, humans need to trust their robotic counterparts. Understanding how robots can provide explanations that foster human trust is an important step toward enabling humans and robots to work together.
Mark Edmonds, PhD, Candidate in Computer Science, University of California, Los Angeles. Yixin Zhu, Postdoctoral Scholar in Computer Science, University of California, Los Angeles.
This article first appeared on The Conversation.
See the article here:
The next frontier of human-robot relationships is building trust - Scroll.in
- A crewless boat is recreating the Mayflower's 400-year-old journey, with the help of artificial intelligence - CBS News - December 26th, 2020
- Artificial Intelligence (AI): Is It All Just Costly Hype? - Dice Insights - December 26th, 2020
- 7 Types of Artificial Intelligence: Propelling the Technology Development - Analytics Insight - December 26th, 2020
- Artificial intelligence will go mainstream in 2021 The Manila Times - The Manila Times - December 26th, 2020
- Top 7 Amazing Artificial Intelligence Applications that Are Unfamiliar - Analytics Insight - December 26th, 2020
- NASCAR Team up with the World Poker Tour while Artificial Intelligence Breaks New Barriers - HighstakesDB - December 26th, 2020
- Youve heard of Elf on the Shelf; in New Haven theres elf with artificial intelligence - New Haven Register - December 26th, 2020
- How is Artificial Intelligence Transforming the lives of People with Disabilities? - Analytics Insight - December 26th, 2020
- How Bad Is Your Spotify? There's Artificial Intelligence to Judge Your Taste - Distractify - December 26th, 2020
- Artificial intelligence measure included in must-pass government funding deal - Homeland Preparedness News - December 26th, 2020
- PetrSU joined the fight for leadership in the field of artificial intelligence applications in Russia and in the world - India Education Diary - December 26th, 2020
- Why not enlist the help of artificial intelligence to nudge people to mask up? - Houston Chronicle - December 22nd, 2020
- Congress wants an inventory of all AI projects at the Pentagon - C4ISRNet - December 22nd, 2020
- The influence of artificial intelligence on the banking industry - Latest Digital Transformation Trends | Cloud News - Wire19 - December 22nd, 2020
- Application of artificial intelligence in the field of treatment and medicine - Tehran Times - December 22nd, 2020
- Using Artificial Intelligence To Transform An Industry? Nominations For The 2021 AI 50 List Are Open - Forbes - December 22nd, 2020
- Delivering Proactive Mental Healthcare With Artificial Intelligence - HealthITAnalytics.com - December 22nd, 2020
- Artificial intelligence is intertwined in precision agriculture - AGDAILY - December 22nd, 2020
- Spin Announces Cutting Edge On-Vehicle Artificial Intelligence (AI) Platform To Bring Sidewalk Riding and Parking Detection Technology to Cities... - December 22nd, 2020
- UC convenes Artificial Intelligence Working Group to harness innovative technology, establish guardrails for equitable and ethical use - University of... - December 22nd, 2020
- Automotive Artificial Intelligence Market Research Report by Technology, by Process, by Offerings, by Application - Global Forecast to 2025 -... - December 22nd, 2020
- Artificial Intelligence, Contactless Solutions Set to Provide Safety Protocols to Enable Reopening of Live Events in 2021 - PRNewswire - December 22nd, 2020
- Employing artificial intelligence to spot infrastructure issues - Construction Dive - December 22nd, 2020
- CPPI and HSDF Release Report from Virtual Symposium: Artificial Intelligence: Transforming the Government Mission - PRNewswire - December 22nd, 2020
- Top 10 AI and machine learning stories of 2020 - Healthcare IT News - December 17th, 2020
- The 8 Best Books About Artificial Intelligence to Read Now - WIRED - December 17th, 2020
- KODA AUTO wins Pioneer Award for its use of artificial intelligence and Sound Analyser app - Automotive World - December 17th, 2020
- Artificial Intelligences Are Power Hungry, But Not How You Think - Nextgov - December 17th, 2020
- Now is the time to ensure artificial intelligence works for Europeans - EU News - December 17th, 2020
- Currux Vision LLC Announces Industry Leading Accuracy of Artificial Intelligence Smart City Traffic Platform Testing with the City of San Jos - PR Web - December 17th, 2020
- Leveraging artificial intelligence to reach fathers and support child nutrition in India during COVID-19 - World Bank Group - December 17th, 2020
- How artificial intelligence helped me overcome my dyslexia - The Guardian - December 17th, 2020
- Artificial Intelligence Advances Showcased at the Virtual 2020 AACC Annual Scientific Meeting Could Help to Integrate This Technology Into Everyday... - December 17th, 2020
- Global Artificial Intelligence in Supply Chain Management Market was Valued at US$ 1549.5 Mn in 2019 Growing at a CAGR of 25.12% over the Forecast... - December 17th, 2020
- Are AI and job automation good for society? Globally, views are mixed - Pew Research Center - December 17th, 2020
- Landmark artificial intelligence legislation advances toward becoming law Defense bill awaits possible presidential veto and Congressional override;... - December 17th, 2020
- The Next Generation Of Artificial Intelligence - Forbes - October 16th, 2020
- St. Louis Is Grappling With Artificial Intelligence's Promise And Potential Peril - St. Louis Public Radio - October 16th, 2020
- Looper column: Artificial intelligence, humanity and the future - SouthCoastToday.com - October 16th, 2020
- For artificial intelligence to flourish, governments need to think ahead about its responsible use - Firstpost - October 16th, 2020
- MediFind Selected as Finalist for Best in Artificial Intelligence in the Shorty Social Good Awards - The Wellsboro Gazette - October 16th, 2020
- The relentless threat of artificial intelligence taking our jobs away - Mint - October 16th, 2020
- Will artificial intelligence technology change the dairy industry - positivelyosceola.com - October 16th, 2020
- Penn Medicine researchers use artificial intelligence to identify early signs of Alzheimer's Disease - Express Computer - October 16th, 2020
- Futurism Reinforces Its Next-Gen Business Commerce Platform With Advanced Machine Learning and Artificial Intelligence Capabilities - Yahoo Finance - October 16th, 2020
- SparkCognition Advances the Science of Artificial Intelligence with 85 Patents - PRNewswire - October 16th, 2020
- Dont Be Afraid, BMW Promises To Keep Artificial Intelligence (AI) On A Tight Leash With These 7 Principles - CarScoops - October 16th, 2020
- Artificial Intelligence in Government: Global Markets 2020-2025 - ResearchAndMarkets.com - Business Wire - August 31st, 2020
- Import Screening Pilot Unleashes the Power of Data - FDA.gov - August 31st, 2020
- Supercharge vegetation management and outage prediction with artificial intelligence - Utility Dive - August 31st, 2020
- Automation, artificial intelligence to be central in the post-Covid world - Economic Times - August 31st, 2020
- Artificial Intelligence Is the Next Top Gun - Bloomberg - August 31st, 2020
- Defense Innovation Unit Teaching Artificial Intelligence to Detect Cancer - Department of Defense - August 31st, 2020
- Toward a machine learning model that can reason about everyday actions - MIT News - August 31st, 2020
- How Artificial Intelligence is changing the way we search and find online - ITProPortal - August 31st, 2020
- TikTok Assets Cant Be Sold Without Chinas Approval - Bloomberg - August 31st, 2020
- Artificial Intelligence in Oil and Gas Market Global Size, Growth and Demand 2020 to 2030 - The News Brok - August 31st, 2020
- Rage Against the Algorithm: the Risks of Overestimating Military Artificial Intelligence - Chatham House - August 31st, 2020
- U of I to lead two of seven new national artificial intelligence institutes - University of Illinois News - August 31st, 2020
- Holocaust survivors will be able to share their stories after death thanks to a new project - 60 Minutes - CBS News - August 31st, 2020
- Funding boost for artificial intelligence in NHS to speed up diagnosis of deadly diseases - GOV.UK - August 31st, 2020
- Artificial Intelligence Identifies 80,000 Spiral Galaxies Promises More Astronomical Discoveries in the Future - SciTechDaily - August 31st, 2020
- Is Artificial intelligence the Future of IT Help Desk? - Analytics Insight - August 10th, 2020
- Experts Are Divided Over Future Of Artificial Intelligence But Agree On Its Growing Impact - Outlook India - August 10th, 2020
- "Artificial Intelligence Will Make Medicine Better in the Long Run" - Biophotonics.World - August 10th, 2020
- How machine learning and artificial intelligence can drive clinical innovation - PharmaLive - August 10th, 2020
- Top Artificial Intelligence and Robotics Investments in July 2020 - Analytics Insight - August 10th, 2020
- Job interviews: Recruiters are using artificial intelligence to analyse what you say to find the right hire - TechRepublic - August 10th, 2020
- Game-Changing Artificial Intelligence Solution by PhotoShelter to Revolutionize Social Media Workflow as Premier Lacrosse League Returns to the Field... - August 8th, 2020
- How Artificial Intelligence is Helping to Fight against Coronavirus in India? - Analytics Insight - August 8th, 2020
- BLOG: How to capitalise on the Artificial Intelligence theme - Your Money - Your Money - August 8th, 2020
- Artificial Intelligence and Its Partners - Modern Diplomacy - August 8th, 2020
- Artificial intelligence isnt destroying jobs, its making them more inclusive - The Globe and Mail - August 8th, 2020
- RadNet and Hologic Announce Collaboration to Advance the Development of Artificial Intelligence Tools in Breast Health - GlobeNewswire - August 7th, 2020
- COVID-19 Impacts: Artificial Intelligence-as-a-Service (AIaaS) Market Will Accelerate at a CAGR of Over 48% Through 2020-2024|Growing Adoption of... - August 7th, 2020
- Artificial Intelligence in Business: The New Normal in Testing Times - Analytics Insight - August 7th, 2020
- VIEW: Digitisation in pathology and the promise of artificial intelligence - CNBCTV18 - August 7th, 2020
- What It Means to Be Human in the Age of Artificial Intelligence - Medium - August 7th, 2020
- Not so Artificial Intelligence When is AI really AI? - EFTM - August 3rd, 2020
- IMD to explore artificial intelligence to improve forecasting, predict extreme weather events - Firstpost - August 3rd, 2020