This AI article reveals a new approach to understanding deep learning models: unraveling the “where” and “what” with concept relevance propagation (CRP)

The field of Machine Learning and artificial intelligence has become very important. We have new advances that have been there every day. The area is having an impact in all areas. By using finely developed neural network architectures, we have models that are distinguished by extraordinary accuracy within their respective sectors.

Despite their precise performance, we are yet to fully understand how these neural networks work. We must know the mechanisms that govern the selection and prediction of attributes within these models to observe and interpret the results.

The intricate and nonlinear nature of deep neural networks (DNNs) often leads to conclusions that may show a bias toward unwanted or undesirable traits. The inherent opacity of their reasoning poses a challenge, making it difficult to apply machine learning models in various relevant application domains. It is not easy to understand how an ai system makes its decisions.

Accordingly, Prof. Thomas Wiegand (Fraunhofer HHI, BIFOLD), Prof. Wojciech Samek (Fraunhofer HHI, BIFOLD) and Dr. Sebastian Lapuschkin (Fraunhofer HHI) introduced the concept of relevance propagation (CRP) in their article. This innovative method offers a path from attribution maps to human-comprehensible explanations, enabling individual ai decisions to be elucidated through human-comprehensible concepts.

They highlight CRP as an advanced explanatory method for deep neural networks to complement and enrich existing explanatory models. By integrating local and global perspectives, CRP addresses the “where” and “what” questions about individual predictions. The ai ​​ideas used by CRP, their spatial representation in the input, and the individual segments of the neural network responsible for their consideration are revealed by CRP, in addition to the relevant input variables that affect the choice.

As a result, CRP describes the decisions made by ai in terms that people can understand.

The researchers emphasize that this explainability approach examines the entire prediction process of an ai, from input to output. The research group has already created techniques to use heat maps to demonstrate how ai algorithms make judgments.

Dr. Sebastian Lapuschkin, head of the Explainable artificial intelligence research group at Fraunhofer HHI, explains the new technique in more detail. He said that CRP transfers the explanation from the input space, where the image with all its pixels is located, to the semantically enriched conceptual space formed by higher layers of the neural network.

The researchers further said that the next phase of ai explainability, known as CRP, opens up a world of new opportunities to investigate, evaluate and improve the performance of ai models.

Insights into the representation and composition of ideas within the model and a quantitative assessment of their influence on predictions can be gained by exploring model designs and application domains using CRP-based studies. These investigations harness the power of CRP to delve into the intricate layers of the model, unraveling the conceptual landscape and evaluating the quantitative impact of various ideas on predictive outcomes.

Review the Paper. All credit for this research goes to the researchers of this project. Also, don’t forget to join. our 31k+ ML SubReddit, Facebook community of more than 40,000 people, Discord channel, and Electronic newsletterwhere we share the latest news on ai research, interesting ai projects and more.

If you like our work, you’ll love our newsletter.

We are also on WhatsApp. Join our ai channel on Whatsapp.

<!– ai CONTENT END 2 –>