As machine learning-based applications reach the hands of operators, the models often struggle to communicate the confidence of their predictions. Operators may end up blindly trusting a model, unaware that the model is only marginally confident or has never seen the input data during training. We present a visualization dashboard for encrypted network traffic labels that combines confidence sliders and visualizations to contextualize the model’s uncertainty.
All the documents including the poster, abstract, videos, and descriptive text have this distribution statement:DISTRIBUTION STATEMENT A. Approved for public release. Distribution is unlimited. This material is based upon work supported under Air Force Contract No. FA8702-15-D-0001. Any opinions, findings, conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the U.S. Air Force. © 2021 Massachusetts Institute of Technology. Delivered to the U.S. Government with Unlimited Rights, as defined in DFARS Part 252.227-7013 or 7014 (Feb 2014). Notwithstanding any copyright notice, U.S. Government rights in this work are defined by DFARS 252.227-7013 or DFARS 252.227-7014 as detailed above. Use of this work other than as specifically authorized by the U.S. Government may violate any copyrights that exist in this work.