JHonaker

17 ms·

121.
by JHonaker 4y ago
Constraint propagation and search are hardly new techniques. In fact they're both from or from before the previous "AI winter." It's a shame they've been left out of so many of the deep learning generation's wh
122.
by JHonaker 4y ago
I’m still not sure what you mean by “safe.” Are you worried about Racket becoming unmaintained? Racket itself is written on top of Chez Scheme, so does that satisfy your safety condition?
123.
by JHonaker 5y ago
What exactly do you mean by this question? R7RS could totally be used to write a library that allows running any racket/base program. It's called an interpreter :D. The nice thing about Racket is the #lang system allows you to wor
124.
by JHonaker 5y ago
I really hope that's a very odd joke. This makes me extremely sad for this person.
125.
by JHonaker 5y ago
If you're interested in a lisp for scripting, you should give Racket a try. This is just a Racket DSL (created I believe as a make-like system for the actual Racket project). Regular racket has all of the things, and if easy interactio
126.
by JHonaker 5y ago
The worst part about Emacs is the bindings. :D The reason I'll never leave Emacs is that if I want to change something, I can. It doesn't really matter what it is. This might be technically true in something like VS Code (although
127.
by JHonaker 5y ago
I mapped Control to a key under my thumb instead of Caps Lock. That's where it was originally when they chose the binding anyway. It's very comfortable now.
128.
by JHonaker 5y ago
I got frustrated with how much other things were designed around Qwerty. Plus I occasionally had to help others on a different computer, and it was always a weird thing to spam the wrong keys for a minute or two while my brain switched back
129.
by JHonaker 5y ago
I’ve switched to Dvorak and Colemak before. I always come crawling back to Qwerty. The reality is the benefits you get from switching keyboard layouts is marginal at best. The real benefits (in terms of RSI) come from things like a split ke
130.
by JHonaker 5y ago
I started this expecting to be disappointed, but I really like all of the principles they're describing. I've been pushing for more of this attitude at my own company.
131.
by JHonaker 5y ago
Just a note in case someone goes looking for racket-xp-mode explicitly. It's just an ancillary mode included with racket-mode. You'll pretty much always want it turned on. I'm sure there's some reason to turn it off some
132.
by JHonaker 5y ago
There is an obvious difference though. As an individual, you provide a small amount to a collective pool kept by a third-party. You can't do much with the effective amount, but the combined wealth of all of an insurers premiums is quit
133.
by JHonaker 5y ago
Not that I'm knocking the deep learning aspects of the field, but what are the interesting non-DL avenues currently being explored?
134.
by JHonaker 5y ago
I will preface this with, "I absolutely love SICP", but I think there are better places to go for everything it teaches. In terms of breadth and information per page ratio there's almost nothing better though. - For how to ac
135.
by JHonaker 5y ago
It's striking how many of the responses are just trying to flex their knowledge of FP. The real answer is that Common Lisp is not a pure functional language. Many CL idioms involve mutation, the most obvious one being setf. So, it mana
136.
by JHonaker 5y ago
I’m a big fan of Racket. It’s basically a batteries included Scheme (used to be called PLT Scheme) with an extra focus on ease of development of DSLs. As an example, do you want a statically typed Scheme? Just change the language directive
137.
by JHonaker 5y ago
Yea, I have absolutely no issues when running a mainstream distro with mostly default behavior (PopOS/Ubuntu). When I veered off into StumpWM land, I had a really strange issue that I had to debug. My laptop is pretty much always in &q
138.
by JHonaker 5y ago
I highly doubt that a person with a PhD in statistics doesn't identify as a statistician.
139.
by JHonaker 5y ago
Is there some sort of special mechanism for partial application in APL that isn’t just defining a partial operator, like a macro or function? It’s not too hard to do yourself, and I’m sure someone’s already written it. I wouldn’t be surpris
140.
by JHonaker 6y ago
Not only that, but the LTSpice solution was incorrect, and the Julia solution was correct.
141.
by JHonaker 6y ago
I do that too, but out of habit from other pursuits rather than a real necessity. Recently, I've found myself going to Emacs and packages' internal help first, and I usually find what I'm looking for. Despite me spending seve
142.
by JHonaker 6y ago
There's a mix of GPL-2, GPL-3, and MIT.
143.
by JHonaker 6y ago
I actually use Emacs to author R Markdown and the exporting works pretty flawlessly. Before anyone asks, I use Org Mode. I just don't feel like pushing that format on my co-workers. If it's for me only, I use Org. If it's for
144.
by JHonaker 6y ago
It's perfectly fine, theoretically for the Markov chain to depend on more than the previous token. When you start to do that though, unless you can write down a function (or approximation, like an NN) the number of states in your state
145.
by JHonaker 6y ago
Also, as you can see. It makes me a terrible editor. I missed several parts where I'd taken out the part introducing my wife into the story, and I made a quite a few silly mistakes. It should read " un diagnosed ADHD" not &qu
146.
by JHonaker 6y ago
I didn't get diagnosed until my late 20s, so I feel like I can articulate some of the differences. One of the things that contributed to that is that I am incredibly lucky in that math, statistics, and computer science are the things t
147.
by JHonaker 6y ago
That’s really interesting. Thanks for sharing.
148.
by JHonaker 6y ago
Sorry, I guess I was unclear. It's definitely a goal for the AD ecosystem, but, as far as I'm aware, AD is not a design goal of Julia itself. That was the point I was trying to make.
149.
by JHonaker 6y ago
I was unsure if every ODE solver was able to be differentiated through. The "some" was me trying to avoid making a blanket statement about all, when I wasn't sure. Sorry for the confusion.
150.
by JHonaker 6y ago
I kind of agree with the author's major sentiment: that ML research is stuck in a rut with incremental improvement. However, the longer the article goes on, the less and less I agree with any of their statements. They start of criticiz

More ›