Artificial and Human Intelligence with Digital Twins
Figure 7: The Average Temperature and Average Action of the RL Algorithm
settings. This will greatly help cases where
research is needed on the parameter
settings. 15 16
Hyperparameter tuning
For
all
deep
learning
methods,
hyperparameter tuning is an important step.
Hyperparameter
settings
are
often
dependent on the domain knowledge of the
application. Research into the specific
application can yield a set of parameter
settings to be tested. In some cases, a set of
parameter settings has been established as
best practices. In other cases, research will
be needed to determine the best settings.
Computer vision
Computer vision is a powerful tool that has
caught the attention of many with its ability
to recognize faces and objects within a
scene. For digital twins, it can add important
information about the quality of the things
being monitored. A task that requires visual
inspection could be enhanced with an AR
interface to a digital twin. For example,
computer vision can detect defects by
comparing thousands of images for
anomalies that may not be as easily detected
by a human. Moreover, specialized cameras,
such as infrared, allow for even further
One feature in SAS ® Visual Data Mining and
Machine Learning is hyperparameter
autotune. This feature will take a range of
potential parameter settings and perform an
optimal search for the best performing
15
Koch, Patrick, et al. 2018. “Autotune: A Derivative-Free Optimization Framework for Hyperparameter Tuning.” Available:
https://www.kdd.org/kdd2018/accepted-papers/view/autotune-a-derivative-free-optimization-framework-for-
hyperparameter-tuning
16
Koch, Patrick, Brett Wujek, and Oleg Golovidov. 2018. “Managing the Expense of Hyperparameter Autotuning.” Proceedings
of the SAS Global Forum 2018 Conference. Cary, NC: SAS Institute Inc. Available:
https://www.sas.com/content/dam/SAS/support/en/sas-global-forum-proceedings/2018/1941-2018.pdf
- 68 -
November 2019