In the Spotlight – Meet Nava Tintarev

In the Spotlight - Meet Nava Tintarev

At Women in AI we’d like to inspire others by featuring female role models that are making a difference in AI. We aim to empower women who are working in AI by highlighting their success stories. This way we hope to inspire other women and girls to get into STEM-related fields. In our series ‘In the Spotlight’ we shine a light on an expert in the field and today we’d like you to meet Nava Tintarev.

Author: Ingrid van Heuven van Staereling.

Who is Nava Tintarev?

Nava is a full Professor in Explainable Artificial Intelligence. She is currently active at Maastricht University in the Department of Advanced Computing Sciences (DACS) as a Director of Research. At the same time she is also a member of the Explainable AI (XAI) research theme. “This is an interdisciplinary field of research focusing on providing explanations of AI-driven systems.” – Nava explains. From intrinsically explainable approaches to AI to methods that provide explanations for decisions made by ‘black-box’ machine learning models: Nava has seen it all. “We bring together researchers working on XAI in a diverse range of application areas, including recommender systems, computational social science, causal inference for life sciences, affective computing, computer vision, knowledge representation/reasoning and machine learning.” -she says. This incredibly important research helps to optimize the ability of AI-driven systems to explain their decisions in human-understandable terms. “This improves the trust we place in these systems and helps address issues related to fair and unbiased decision-making.”

Making AI Trustworthy

Decision-making at individual, business, and societal levels is influenced by online content. And filtering and ranking of algorithms are used to support these decisions. However, the design of many of these systems may lead to information silos and filter bubbles. This is where Nava and her team of researchers step in. “In my work, I use algorithms that help select information from diverse viewpoints. Diversification is, however, likely to result in content that is surprising to users. If it is not clear to a user why surprising content is shown, it can be difficult for them to trust the digital curation that occurs.” Nava considers this when developing models for the design of interactive interfaces that support people in both understanding and adjusting filtering and ranking mechanisms in a way that is meaningful to them. Such interactive interfaces can help users understand and change how filtering and ranking occur. A useful way to bring to light what might not have been made clear. “The design choices applied to these interfaces also serve to encourage the kind of critical thinking that inoculates readers against future misleading information.” – Nava tells us. This will help the users of these ranking mechanisms have a better understanding of why and how this ranking has taken place. It gives them more control in what they feel is right to use.

“I have worked on projects funded by IBM, Twitter, and an EU Marie-Curie ITN on Interactive Natural Language Technology for Explainable Artificial Intelligence.” At the moment Nava is representing Maastricht University as a Co-Investigator in the ROBUST consortium, selected for a national (NWO) grant with a total budget of 95M (25M from NWO) to carry out long-term (10-years) research into trustworthy artificial intelligence. She is eager to work on this topic of making AI (outcomes) better. This is why she is also a co-director of the TAIM lab on trustworthy media together with RTL and the University of Amsterdam. Additionally to that she also serves on the advisory board for a Strategic Research Program (SRP) on recommender systems in media markets in Belgium.

“To create a more tolerant society, we need to be aware of different views and understand that resolving differences is part of the democratic deliberative process."

A Human-Understandable Approach

Nava’s research considers how she can support knowledge goals of individuals such as learning about topics that are similar to what they already know about but still new. But it is also about deepening the knowledge on a specific topic or exploring completely new topics. Nava is convinced this reflects on higher-level societal values, such as tolerance and democracy. “To create a more tolerant society, we need to be aware of different views and understand that resolving differences is part of the democratic deliberative process. Awareness of these values underpins all of the design decisions in my research.” Nava considers things like:

Which types of data are needed to represent diversity of viewpoint
The computational metrics for measuring diversity
Diversity enhancement
The design of an interactive user interface for diverse content presentation

“I focus on developing explanations that work for people and also evaluating them with people.” She does this by working on automatically generated explanations (transparency) and explanation interfaces (recourse and control). All in contexts where explanations are particularly relevant, such as when dealing with different viewpoints online.

Find the Beat of Your Own Drum and Keep Learning

When it comes to working within the field of AI, Nava would like all (young) women that are interested in this field to know that this is not a single career path. “The most exciting jobs are the ones you create along the way. The trick is to keep in tune with the beat of your own drum while staying in touch with the needs of society and the world around you.” Nava is convinced that programming skills are very useful to start with. “You need to be able to understand how software works to know its limitations and abilities well.” But you also need to be able to keep up to date. “The state of the art in our field moves quickly, you need to be interested in keeping up to date by reading about it.” But to make sure you have an impact and are able to grow as a leader, Nava emphasizes that you have to be able to understand and reach others. “There are no shortcuts. Do, read and try! Let curiosity be your guide.” She encourages other women to have patience, and not be afraid to fail. “Each so-called ‘failure’ is an opportunity to learn.”