The emerging field of Explainable AI

Tiếng Việt

The emerging field of Explainable AI

Currently Artificial Intelligence (AI) solutions are ‘black boxes’, since AI decisions lack transparency and their rationale cannot be clearly explained yet, according RMIT academic Dr Dang Pham Thien Duy.

Dr Minh talking to an audience of students, alumni and industry experts.

The emerging field of Explainable AI is a response to this challenge. Efforts are being made around the world to make AI more transparent to its users to address critical questions around ethics, such as who is responsible for biases produced by AI, who AI serves, and how best to deal with privacy concerns relating to the vast amounts of data being collected by businesses and organisations.

Dr Duy, an RMIT Information Technology Lecturer along with RMIT Software Engineering Lecturer Dr Minh Dinh recently hosted an industry panel discussion to shed light on the AI opportunities and challenges that currently exist in Vietnam.

Dr Duy presenting at Demystifying AI: Applications, opportunities and challenges in Vietnam.

Dr Minh described Explainable AI as “a very exciting field of study to engage in”.

“Success in making AI operations more transparent will help us understand what characteristics in our data lead to certain decisions and classifications made by AI, how bias can be identified and mitigated, and how to improve AI performance both in terms of accuracy and processing time,” Dr Duy shared.

Two big AI opportunities in Vietnam identified by Dr Minh and Dr Duy were enhancing cybersecurity and human resources to marketing and health care industries due to the increased adoption of AI and Machine Learning (ML) technology in the local market.

“To successfully engage AI and ML tools for your business, the first and most important factor is the amount of data you can gather. However, there is a challenge in preparing the data for your ML objectives, as data is often in inconsistent formats and consists of a lot of noise and bias,” Dr Minh said.

“Second is expertise in data science and data analytics. This is in fact a great barrier for businesses to successfully devise and deploy AI and ML projects.”

The invited panel speakers (from left): Dr Yossi Nygate (RMIT), Herve Roussel (Quod AI), Thai Dinh (Be Group), Oscar Lopez Alegre (Leroy Merlin Vietnam), and Dr Khanh Ngo (Kaspersky SEA).

The invited panel speakers (from left): Dr Dang Pham Thien Duy (RMIT), Dr Yossi Nygate (RMIT), Herve Roussel (Quod AI), Dr Khanh Ngo (Kaspersky SEA), Thai Dinh (Be Group), Oscar Lopez Alegre (Leroy Merlin Vietnam), and Dr Minh Dinh (RMIT).

The panel discussion Demystifying AI: Applications, opportunities and challenges in Vietnam was organised by RMIT’s School of Science & Technology in December.

The panel also included Quod AI Co-Founder Mr Herve Roussel, Data Science at Be Group Head Mr Thai Dinh, Kaspersky SEA Country Manager Dr Khanh Ngo, Leroy Merlin Vietnam Chief Technology Officer Mr Oscar Lopez Alegre, and RMIT School of Science & Technology Senior Lecturer Dr Yossi Nygate.

Story: Jamila Ahmed

Science and technology

07 February 2020