Ajay Chander

Explainable AI: The New 42?

Explainable AI is not a new field. Since at least the early exploitation of C.S. Pierce’s abductive reasoning in expert systems of the 1980s, there were reasoning architectures to support an explanation function for complex AI systems, including applications in medical diagnosis, complex multi-component design, and reasoning about the real world. So explainability is at least as old as early AI, and a natural consequence of the design of AI systems. While early expert systems consisted of handcrafted knowledge bases that enabled reasoning over narrowly well-defined domains (e.g., INTERNIST, MYCIN), such systems had no learning capabilities and had only primitive uncertainty handling. But the evolution of formal reasoning architectures to incorporate principled probabilistic reasoning helped address the capture and use of uncertain knowledge.

Guided Play: Automatic Stereotypical Behavior Analysis and Intervention during Play

Restricted and repetitive behaviors (RRB) are a core symptom and an early marker of Autism Spectrum Disorder (ASD). Despite technologies for detecting certain forms of RRB, assessment and intervention for RRB still heavily rely on professional experience and effort. This paper presents an ongoing investigation of a technology that uses instrumented games or toys as platforms to assess RRB and facilitate behavior intervention during play. The design and implementation of a prototype for the iPad are discussed. The same technology can be applied to tangible objects such as smart toys for a natural player-computer interface.

Evaluating Explanations by Cognitive Value

The transparent AI initiative has ignited several academic and industrial endeavors and produced some impressive technologies and results thus far. Many state-of-the-art methods provide explanations that mostly target the needs of AI engineers. However, there is very little work on providing explanations that support the needs of business owners, software developers, and consumers who all play significant roles in the service development and use cycle. By considering the overall context in which an explanation is presented, including the role played by the human-in-the-loop, we can hope to craft effective explanations. In this paper, we introduce the notion of the “cognitive value” of an explanation and describe its role in providing effective explanations within a given context. Specifically, we consider the scenario of a business owner seeking to improve sales of their product, and compare explanations provided by some existing interpretable machine learning algorithms (random forests, scalable Bayesian Rules, causal models) in terms of the cognitive value they offer to the business owner. We hope that our work will foster future research in the field of transparent AI to incorporate the cognitive value of explanations in crafting and evaluating explanations.

Last mile end-user programmers: programming exposure, influences, and preferences of the masses

In this paper, we set out to explore the level of programming experience present among the masses (the last mile end-user programmers), the influence of various factors such as early exposure to software, as well as age, on programming experience, their effects on the types of software people mightwant to create, and the software development approaches they prefer.

Exploring the Dynamics of Relationships Between Expressed and Experienced Emotions

Conversational user interfaces (CUIs) are rapidly evolving towards being ubiquitous as human-machine interfaces. Often, CUI backends are powered by a combination of human and machine intelligence, to address queries efficiently. Depending on the type of conversation issue, human-to-human conversations in CUIs (i.e. a human end-user conversing with the human in the CUI backend) could involve varying amounts of emotional content. While some of these emotions could be expressed through the conversation, others are experienced internally within the individual. Understanding the relationship between these two emotion modalities in the end-user could help to analyze and address the conversation issue better. Towards this, we propose an emotion analytic metric that can estimate experienced emotions based on its knowledge about expressed emotions in a user. Our findings point to the possibility of augmenting CUIs with an algorithmically guided emotional sense, which would help in having more effective conversations with end-users.

Enterprise Assistant Service: Supporting Employees in the Digital Enterprise

Enterprises globally are seeking out and leveraging digital technologies to improve their performance and competitiveness. As data-driven personalization becomes an increasingly ubiquitous aspect of our digital experience, we believe it is likely that the rapidly digitizing workplace will explore systems for personalizing the support their employees receive. In this paper, we present our experience designing and experimenting with a pilot service that provided personalized digital tool recommendations to enterprise users, for work-related issues. This Enterprise Assistant service, or EAS, was offered for 10 weeks and served 24 users within the same enterprise. Users emailed the EAS with their questions and received personalized suggestions and follow-ups until their issue was resolved. The service addressed a variety of issues during the experiment, with a majority of users expressing interest in continuing to use it. One key finding is that user awareness of friction points in their daily workflows is quite low, leading to significant communication overhead simply to uncover an actionable issue for the EAS. We channel our findings towards design guidelines and opportunities for systems that aim to empower employees with personalized tools in our rapidly digitizing workplaces.

The Lifeboard: Improving Outcomes via Scarcity Priming

We introduce the Lifeboard: a dynamic information interface designed to render personal data so as to positively influence wellness outcomes. We report on the results of an experiment that compares the effect of presenting clinically significant data to subjects on their activity levels, with the effect of presenting the same data using the Lifeboard. The statistically significant increase in this wellness outcome in the Lifeboard group vs. the Data-only group suggests that the Lifeboard effectively leverages the scarcity response [4] in the service of improved wellness outcomes. Moreover, the significant week-on-week decrease in this wellness outcome in the Data-only group points to the need for care when exposing clinical data to users.