Under a collaboration, we are developing adaptive autopilots to handle rapid unmodeled aircraft configuration changes, Dr. Ankit Goel visits us
...
News
Thu, 11/14/2019
Ziwei Li and Sai Ravela discuss how
- The predictability of neural networks is near identical to the chaotic dynamics they learn from
- This they argue is because neural Networks follow a path to chaos through alternate stretching and compression
- Very few training...
Mon, 11/04/2019
Sai Ravela presents at Shell, arguing for theory-data symbiosis, which includes:
- Models from theory suffer from nonlinearity, dimensionality and uncertainty
- Models from data suffer from the above plus issues of generalization, extrapolation, and the representation of theory...
Fri, 10/04/2019
Michael Barbehenn and Alton Barbehenn
Based on catch data from an open population of animal exposed to a grid of traps for several days, we can accurately & precisely estimate the population density (abundance). We assume animals are terrestrial (not water or air), individual (not...
Fri, 09/20/2019
Goran Zivanovic, PhD Student, EECS, with advisors Seager and Ravela, presents his work on developing generative light-curve models in exoplanet studies. Promising!
Fri, 09/13/2019
Sai Ravela presented his work on production of Exact Neural Networks using Theory, as well as matched spectra as a basis for configuring learning.
Thu, 09/12/2019
Sonia's been working on the dynamics of learning...she got featured!
Fri, 09/06/2019
Ziwei Li presented his work on Learnability of Chaotic Dynamics with application to the Lorenz system. It is amazing to see how quickly the Lorenz system captures the attractor structure.
Fri, 08/16/2019
Kshitij Bakliwal, ESSG Visiting Student 2019-20
Sai Ravela
We present a new algorithm for recognition that uses mass transport distances and amplitude (filtered brightness) errors as inputs of a convolutional neural network that learns to recognize images in Siamese...
- ‹ previous
- 3 of 6
- next ›