hideo7746

Because an excess of stamps can't feed, clothe, or educate someone. It can't help house someone and it can't save someone's life.

Money on the other hand can do much more good not sitting in the bank account of someone who has more of it than they could ever spend. So keeping it in there for the purpose of 'scoring points' is seen by many, myself included, as obscene.

Also, I disagree that collecting money is the only hobby that would receive this criticism. Personally I would also find it obscene to collect insulin, clean water, and residential housing to name a few.

Location: Ann Arbor, Michigan US

Remote: yes

Willing to relocate: depends on location, prefer US west coast

Technologies: Python, sk-learn, TensorFlow, Javascript/TypeScript, React, AWS and GCP

Resume: https://johnholler.com/resume/

email: [email protected]

Summary

Math PhD with research expertise in deep learning, reinforcement learning, and game theory. Most recently I have been building frontend experiences like https://dev.d3513lez266dga.amplifyapp.com/, and serverless apps like https://beunstuck.net/. I want to transition from teaching into a full-time SWE role, I like frontend work but I'm open to working anywhere on the stack.

Is this an attempt at satire? I mean, the lobster actually reacts to imminent death whereas you've chosen to personify inanimate and already dead lettuce. Do you believe they are the same? I am baffled.

I created a simple, low cost alternative to square appointments called Unstuck (beunstuck.me). You can use it for managing virtual appointments (booking page, payment handling, creating event). I use it for my math tutoring business, and am hoping other tutors and content creators can use it to connect with their audience in a different way.

Ah, did not end up getting that domain name and went with something cheaper. The profiles are at booking.beunstuck.me, you can see mine here: https://booking.beunstuck.me/profile/hCTa0hQ2qAa85SYTer2t0in...

When Covid hit last year, all of my math tutoring business moved online. I took that opportunity to try out a number of different platforms and services to handle scheduling and payment. In the end I wasn't satisfied with any of my options, and so I built my own tool using React and Firebase, called Unstuck, that fit my needs. I've continued working on it in recent months, and built it out so that anyone can create an account and start offering one-on-one online appointments. Unstuck lets you create different kinds of services, set your default availability, and creates a booking page for customers to find a time and request an appointment. Payment is collected up front and transferred through a connected Stripe account. Check out the launch post to learn more, read through the graduate student starter kit to get a better understanding of the product, and see how I connect Unstuck to my personal website here: https://johnholler.com/tutoring

thanks for looking!

*she

On one hand you're right, methods like Q-learning are model-free and do not necessarily encode much about state dynamics. The Q-function is a feature (function) of state and while ita may not say much about the model, it does encode the most important aspect of the model in terms of solving the task. Namely, it predicts the accumulated reward conditional on next actions actions. That makes it a somewhat narrow representation of state on its own. But, if you consider an environment that has many reward signals, and you learn Q functions for each, this ensemble of Q functions can consitute a rich representation of state. Depending on what the reward functions are, the associated Q functions may be sufficient to construct a full model. so I guess my point is that the learned quantities in RL encode key aspects of state, and when you expand beyond the single task/single reward RL setting the lines between value and model can become blurred.