(me)Hi, I'm Andy Ganse. I'm an applied physicist and data scientist in the Seattle area, currently enjoying my role as a radar data analyst and machine learning researcher at Echodyne Corporation, an amazing startup delivering a new type of small-size, low-cost, low-power radar to the drone, security, and autonomous automotive industries. I'm also the principal scientist and founder of Anseres Research & Technology LLC, focused on federal and state funded scientific R&D contracts (but presently only use that business for publishing non-radar-related journal papers). I was a senior research physicist doing computational geophysics for 16 years at UW's Applied Physics Laboratory (APL-UW), and have also worked as a data scientist in machine learning applications in another tech startup. Whether in the commercial world or in the academic science world, the focus of my work has been in machine learning, inverse problems, optimization, signal processing, and data analysis. (My PhD concerned inverse problems in ocean acoustics, and at APL-UW my work was in statistical inference and remote sensing via acoustics, electromagnetics, and gravity). Here on research.ganse.org are some of my publicly shareable research results and tools, both in data science topics and in applied physics topics. If you've seen my UW research website in past you'll recognize many of those pages in here too. You can contact me at andy@ganse.org.

[loan fig]Let's implement a random forest classifier from Scikit-Learn to see how well we can predict whether a bank client will have good loan behavior (meaning they won't default or become delinquent) if they are given a new loan. We'll use a public bank transactions and loans dataset from the PKDD99 Challenge conference for the modeling. In the process we'll fit and explore the assumptions made for this model, and learn about some limitations of Scikit-Learn's tree-based models.

[csem fig]The Conductivity Profiler is an instrument for remotely observing estuarine salinity profiles via electromagnetic measurements. Electromagnetic (EM) waves are attenuated in seawater as a function of frequency, and conductivity structure (closely related to salinity structure) in the water can be inferred by combining measurements of EM waves at different frequencies on a distant electric field receiver. Geophysical inversion methods are applied to estimate the estuarine salinity profile from the EM measurements. Using inverse theory techniques, we take advantage of statistical rigor and let the data determine the structure of the conductivity profile and quantify the uncertainty and resolution of the salinity profile.

[mplr fig]Code to fit multiple co-joined straight lines to a set of data points. Note the popular name (above, or sometimes "segmented linear regression") for this topic may be a bit misleading, as this is in fact a nonlinear regression problem due to solving for the intersection points of the co-joined lines.

[Rpi fig]The tiny version of the Raspberry Pi computer, the Pi Zero, is barely larger than a stick of gum yet is a full Linux-based computer. It's a bit less powerful than the larger Pi 2 & 3 models, but it uses much less power than they do. It's a great field sensor platform for some applications, and super quick to implement in all sorts of projects. Projects here include: 1.) adding a canned sensor attachment and running a Jupyter server on the Pi to analyze/view the sensor data in a web-based Python notebook; 2.) attaching a USB mic and audio amplifier/speaker add-on and implementing the Jasper speech/voice-control agent; 3.) attaching a near-infrared camera and taking nighttime photos in our back yard after dark.

[gps/map fig]Did you know you can plot your geographic data on interactive maps embedded directly in your Python notebooks? Check it out, as we play with and analyze some GPS tracking data. A database of tracked walking routes data available on a health/fitness website provides a convenient trove of data not only to play with, but also to explore the geometric interference effects of downtown buildings upon GPS track solutions.

[NASA fig]
The nature of an icy satellite's interior relates fundamentally to its composition, thermal structure, formation and evolution history, and prospects for supporting life. Gravity measurements via radio Doppler information during spacecraft flybys are an important tool used to infer gross interior structure of these moons. Liquid water and ice layers have previously been inferred for the interiors of Jupiter's icy satellites Europa, Ganymede, and Callisto on the basis of magnetic field measurements by the Galileo probe, and on Europa and Callisto induced magnetic field signatures measured by the Galileo probe provided strong evidence for an ionic aqueous ocean. We apply geophysical inverse theory tools to assess the icy moon's interior density anomaly distribution that could be estimated from radio Doppler measurements, to support the search for mass anomalies in the ice shell (meteorites or diapiric upwellings) or near the H2O/rock interface (seamounts).

[me at sea]
In 2009 and 2010 I participated in a major ocean acoustics experiment in the Philippine Sea with APL-UW's North Pacific Acoustic Laboratory (NPAL) group. One of the key topics our research group is interested in is how oceanic sound propagation is affected by internal waves (waves down deep in the water) and by ocean "spice" (blobs of water with a different soundspeed but same density as their surroundings, so they don't behave like waves). Both these phenomena cause variations in soundspeed of the water and thus acoustic transmissions through it, and acousticians would like to understand them better. Our group is also interested in the estimation of ocean sound-speeds and temperatures from receptions of sound that we send through the water. Find out more about our use of these different methodologies in our research on my NPAL Ocean Acoustics page.

[InvGN fig]Calculate Tikhonov-regularized, Gauss-Newton nonlinear iterated inversion to solve the damped nonlinear least squares problem, using the InvGN toolkit for Matlab/Octave. While Matlab's optimization toolbox contains lsqnonlin, it does not inherently include the Tikhonov regularization and is only the optimization component, whereas inversion also requires uncertainty quantification. My InvGN package is for weakly nonlinear, Tikhonov-regularized, inverse problems (so includes solution uncertainties and resolution quantification), handles both frequentist and Bayesian frameworks, and also works in Octave.