A New Approach to Understanding How Machines Think

4/21/2019

This article introduces Been Kim, a computer scientist and researcher who is working on developing interpretable and transparent artificial intelligence (AI) models. Kim's focus is on creating tools that can help humans better understand how AI systems make decisions. She aims to bridge the gap between the 'black box' nature of many AI models and human comprehension. The article highlights Kim's work on 'model cards', which are like nutrition labels for AI systems, providing details on their limitations, potential biases, and overall performance. Through her research, Kim seeks to build trust and accountability in AI systems by making them more explainable and interpretable.

artificial intelligence ꞏ machine learning ꞏ deep learning ꞏ neural networks ꞏ natural language processing ꞏ ai research ꞏ human-computer interaction ꞏ data visualization ꞏ interpretability ꞏ explainability ꞏ model transparency ꞏ ai ethics ꞏ ai algorithms ꞏ ai applications ꞏ ai technology ꞏ ai systems ꞏ ai development ꞏ ai innovation ꞏ ai programming ꞏ ai frameworks ꞏ ai tools ꞏ ai interpretation ꞏ ai communication ꞏ ai advancements ꞏ ai models ꞏ ai training ꞏ ai decision-making ꞏ ai decision support ꞏ ai integration ꞏ ai automation ꞏ ai translation ꞏ ai language processing ꞏ ai interfaces ꞏ ai user experience ꞏ ai design ꞏ ai implementation