ai_leadership

Marjory Da Costa Abreu - Demystifying Artificial Intelligence

Decoding Artificial Intelligence: A Comprehensive Exploration by Mari About the Expert Originally from Brazil, Mari is a senior lecturer in Applied Artificial Intelligence at Sheffield Hallam University, United Kingdom. Her work in the academic setting spans nearly two decades with a research focus

Decoding Artificial Intelligence: A Comprehensive Exploration by Mari

About the Expert

Originally from Brazil, Mari is a senior lecturer in Applied Artificial Intelligence at Sheffield Hallam University, United Kingdom. Her work in the academic setting spans nearly two decades with a research focus on developing and evaluating artificial intelligence (AI) algorithms.

New Era of Artificial Intelligence

In recent times, an enhanced understanding of intelligent systems and the application of artificial intelligence has emerged. This shift is largely thanks to the boon of big data and improved processing power of computers. However, it's essential to understand that such development of AI systems has a beneficial as well as detrimental impact.

The Bias in Big Data

Large volumes of data used in AI algorithms often contain biases. For instance, real data sets, especially those pertaining to demographics, include inherent biases that end up in the final AI system. Therefore, the popularization and misuse of artificial intelligence have both positive and negative implications.

Understanding Artificial Intelligence, Machine Learning and Deep Learning

It's common to mistake artificial intelligence, machine learning, and deep learning for one another. Realize that artificial intelligence, which represents anything intelligent, encompasses machine learning. Inside machine learning resides deep learning, which has only become possible and popular due to the development around big data.

The Misuse of Black Box in Machine Learning

As AI tools have become more accessible, many users conveniently rely on black box systems. They often use default settings without understanding the results. However, completely relying on default settings can lead to reinforcing biases, sub-optimal results, and a lack of true value from the models. It is important to fine-tune techniques and understand the results deeply.

Neural Networks and Understanding Your Data

Before utilizing a machine learning model, users must understand their data and always be aware of its potential biases. Furthermore, using techniques without proper understanding can lead to misuse, such as not fine-tuning parameters properly in neural networks.

Demystifying the Use of AI

While AI is a powerful tool and can be an effective solution for many problems, it is not a panacea that can solve every issue. Understanding your data, learning the basics of machine learning, and using AI models thoughtfully and responsibly is vital.

Contact Information

Mari welcomes discussions on the ethical use of AI, theoretical AI, machine learning, and more. You can find her on Instagram, LinkedIn, Facebook, and Twitter at ProfMaji.

In conclusion, while AI holds immense potential, it's critical to apply it responsibly and understand the implications fully. With a careful approach, the misuse of AI can be prevented, effectively harnessing its power for the benefit of all.