Visualizing Deep Neural Networks with Topographic Activation Maps

Machine Learning with Deep Neural Networks (DNNs) has become a successful tool in solving tasks across various fields of application. However, the complexity DNNs makes it difficult to understand how they solve their learned task. To improve explainability DNNs, we adapt methods from neuroscience that analyze complex and opaque systems. Here, draw inspiration uses topographic maps visualize brain activity. also activations neurons as maps, research techniques layout two-dimensional space such similar activity are vicinity each other. In this work, introduce compare obtain DNN layer. Moreover, demonstrate use activation identify errors or encoded biases training processes. Our novel visualization technique improves transparency DNN-based decision-making systems is interpretable without expert knowledge Learning.