husband | son | brother | statistician masquerading as a computer scientist | mathematical optimizer masquerading as a statistician | locomotor | board-rider | meditator | provider for šŸ• āˆ§ šŸˆā€ā¬› āˆ§ šŸ“ | lifetime learner | kšŸ„·šŸ¼

OneChronos

Georgia Tech

Biography

I’m Jake, a Nebraska-native-turned-Colorado-convert who’s also moonlighted in Oregon, Utah, and Alaska. I’m married with no human kids—two dogs, a cat, and an adult-aged, lives-in-her-own-place-but-is-still-on-our-dime horse (just kidding, Minnie—you’re great). I’m a middle child with two wonderful sisters. And since it’s often a point of confusion, you pronouce the “K” in my last name, just like you do with Donald Knuth.

I say that I work in finance (but I really work in tech) with interests spanning optimization, statistics, control and reinforcement learning, and (algorithmic) game theory. I’m very fortunate to be part of the OneChronos family (since late 2020); we get to research and engineer solutions to thorny problems at the intersection of machine learning, game theory, optimization, and theoretical computer science (at the speed and scale of electronic trading).

I recently completed Georgia Tech’s online master of science in computer science (where I currently TA CS 7642: Reinforcement Learning and Decision Making). I studied quantitative finance and actuarial studies (aka, finance + applied math) at the University of Colorado for my undergrad.

Before joining OneChronos, I worked with a great cast of characters at CoBank ACB (2015–2020), primarily doing stress testing and credit loss modeling; and prior to that, I was part of a small, tightly-knit band of credit risk modelers and goverance folks at Deloitte (2013–2015). I’ve had the good fortune of working with good people all throughout my life.

In a prior life, I used to be a sponsored snowboarder. I was bestowed the nickname of Kninja (kšŸ„·šŸ¼) by Jesse Burtner for various of my on- and off-snow antics—lots of fun times. The snowboard industry is chock-full of inspiring, creative, good humans.

Outside of work, school and all that, I enjoy spending time with my wife, our families, friends and pets, being outside, using quantitative tools to tell stories and solve problems, (trail) running, learning new things (like languages), and playing stringed instruments (like guitar, banjo and mandolin) rather poorly.

Though still a Molli by surname, Angela (my wife and better half) has the other half of the knigge.us domain. Check it out: Angela’s place! We currently call Fort Collins, CO home, but are relocating to College Station, TX for Angela’s one-year small animal rotating internship at Texas A&M—yeehaw! (Angela’s vet med adventure has taken us from šŸ„¶ (AK) to šŸ” (CO) to šŸ„µ (TX). I think we need šŸŒŠ in the mix too…)

Interests

  • distribution-free statistics
  • high-dimensional probability and statistics
  • (convex) optimization
  • online learning and sequential prediction
  • concentration of measure
  • conformal prediction and inference
  • randomized algorithms

Education

  • Masters - Computer Science (Machine Learning), 2022

    Georgia Tech

  • Bachelors - Quantitative Finance and Actuarial Studies, 2013

    University of Colorado Boulder

Recent Posts

OMSCS debrief - part 1

ISYE 6420: Bayesian Statistics For my first course in the OMSCS program, I decided to take Bayesian Statistics: it had open “seats” (and getting in to classes can be challenging—especially when you don’t have many credit hours);1 I’d self-studied Bayesian statistics a good amount, so I felt confident that I could do well in the class; and from the OMSCS Central reviews, it didn’t seem like it would be an overwhelming first course (in terms of time or difficulty).
2023-02-17 4 min read

OMSCS debrief - part 0

Online Master of Science in Computer Science I’m at the tail-end of Georgia Tech’s Online Master of Science in Computer Science (OMSCS) program—tailend as in, I took my last exam yesterday (at the time of writing this). So in the spirit of revitalizing this blog, I’ve decided to unpack my experience completing the program. Tl;dr: OMSCS has been a good ride; I’m grateful to have had the opportunity to work through it.
2022-05-02 10 min read

LPs and MDPs

(I originally wrote this post in response to a student’s question for Georgia Tech’s CS 7642 Reinforcement Learning and Decision Making course.) Solving MDPs via linear programming Markov decision processes (MDPs) arise in control and reinforcement learning (RL) problems. Modulo notational and cultural differences, control and RL study many similar problems. I highly recommend checking out Ben Recht’s blog series An Outsider’s Tour of Reinforcement Learning for some quality commentary on the control versus RL divide.
2021-08-08 5 min read

The (math behind the) match

My wife’s in vet school and nearing her final year—although the third year really just flows into the fourth—no summer break before rotations! Anyways, in their fourth year, many vet students apply to internships through the American Association of Veterinary Clinicians' Veterinary Internship and Residency Matching Program (VIRMP). Like its human doctor counterpart, the VIRMP aims to pair off interns/residents with colleges/private practice. Indeed, “The VIRMP follows principles established by the Physicians National Intern and Resident Matching Program” (which is typically dubbed the Match).
2021-04-12 8 min read

Update - rerouting

Back in January, I’d planned to do a series on fast direct solvers. It’s a neat topic that bridges numerical linear algebra, randomized linear algebra and algorithms, and physics and engineering—after all, these are solvers for PDEs. Long story short, I’m going to postpone that series for a bit. In the nearer term, I have a few topics that I’d like to unpack, including primers on semidefinite programming (potentially with a foray into sums-of-squares), and game theory and mechanism design (as a related bundle); and a series on common/fundamental trade-offs arising in certain fields, including statistic’s bias versus variance trade-off, reinforcement learning’s exploration versus exploitation trade-off, market design’s stability versus efficiency, and a few others.
2021-04-08 1 min read
See all posts

Popular Topics

article review bag of tricks Bayesians brain exercise concentration of measure convex optimization Efron Fisher Fisherians frequentists fun machine learning numerical analysis online learning optimization probability profiles randomized algorithms statistics updates

Contact