Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Next revision
Previous revision
explanatory [2017/06/29 13:23]
127.0.0.1 external edit
explanatory [2018/04/14 01:12] (current)
admin
Line 301: Line 301:
  
 Deep neural networks (DNNs) have advanced performance on a wide range of complex tasks, rapidly outpacing our understanding of the nature of their solutions. While past work sought to advance our understanding of these models, none has made use of the rich history of problem descriptions,​ theories, and experimental methods developed by cognitive psychologists to study the human mind. To explore the potential value of these tools, we chose a well-established analysis from developmental psychology that explains how children learn word labels for objects, and applied that analysis to DNNs. Using datasets of stimuli inspired by the original cognitive psychology experiments,​ we find that state-of-the-art one shot learning models trained on ImageNet exhibit a similar bias to that observed in humans: they prefer to categorize objects according to shape rather than color. The magnitude of this shape bias varies greatly among architecturally identical, but differently seeded models, and even fluctuates within seeds throughout training, despite nearly equivalent classification performance. These results demonstrate the capability of tools from cognitive psychology for exposing hidden computational properties of DNNs, while concurrently providing us with a computational model for human word learning. Deep neural networks (DNNs) have advanced performance on a wide range of complex tasks, rapidly outpacing our understanding of the nature of their solutions. While past work sought to advance our understanding of these models, none has made use of the rich history of problem descriptions,​ theories, and experimental methods developed by cognitive psychologists to study the human mind. To explore the potential value of these tools, we chose a well-established analysis from developmental psychology that explains how children learn word labels for objects, and applied that analysis to DNNs. Using datasets of stimuli inspired by the original cognitive psychology experiments,​ we find that state-of-the-art one shot learning models trained on ImageNet exhibit a similar bias to that observed in humans: they prefer to categorize objects according to shape rather than color. The magnitude of this shape bias varies greatly among architecturally identical, but differently seeded models, and even fluctuates within seeds throughout training, despite nearly equivalent classification performance. These results demonstrate the capability of tools from cognitive psychology for exposing hidden computational properties of DNNs, while concurrently providing us with a computational model for human word learning.
 +
 +https://​github.com/​slundberg/​shap A unified approach to explain the output of any machine learning model
 +
 +https://​arxiv.org/​abs/​1803.04765 Deep k-Nearest Neighbors: Towards Confident, Interpretable and Robust Deep Learning
 +
 +This hybrid classifier combines the k-nearest neighbors algorithm with representations of the data learned by each layer of the DNN: a test input is compared to its neighboring training points according to the distance that separates them in the representations.
 +
 +https://​arxiv.org/​abs/​1804.03126v1 Data2Vis: Automatic Generation of Data Visualizations Using Sequence-to-Sequence Recurrent Neural Networks
 +
 +