lamchob

Yet, it is usually in middel of the repsective dimensions, not on the extreme ends of either.

Haven't had a chance to read it yet, but I enjoy the authors youtube content (The Engineering Guy): https://www.goodreads.com/book/show/75598048-the-things-we-m...

In the paper (https://dl.acm.org/doi/10.1145/3575693.3575702) there is also a TensorRT comparison. There, they are pretty much neck and neck. Sometimes faster, sometimes slower.

If you read into the paper (https://dl.acm.org/doi/10.1145/3575693.3575702), one can find more performance comparisons. There, from a latency/throughput PoV they are en par with existing tools like TVM/Ansor. Sometimes faster, sometimes slower.

What is more interesting is this: They have very GPU-specific auto-tuning routine that drastically reduces the optimzation space, compared to TVM/Ansor. They go from ~10^6 possible implementations for an operator to a "few hundred", which enabled much faster time-to-solution. This is achieved with a GPU-centric problem formulation and search space. In essence, they trade how widely applicable their approach is (from "any" kind of hardware to only GPU-style architectures) for retrieval speed.

Im currently working on a project, involving a language I‘m not familiar with. It is super simple to paste a function or section of code, have the functionality explained to me and iterate on any open questions.

Last year, a Buell XB12R Firebolt. An exotic, American sport bike with a radical design and engineering philosophy. I knew they existed, but never paid much attention. Then a distant friend sold his bike. I took a test ride, just for the fun of it, because they are quite rare.

On the ride home I already had made the downpayment with the intent to pick up the bike the next day. One the best decisions I ever made!

This reminds me of a movie, the name of which I can’t remember.

A family father gets premonitions about a massive storm that will come, which triggers him building a bunker in his back yard. It is a family drama on how his obsession with protecting his family slowly tears it apart. In the end they convince the father to go on vacation for a few days, just to get away from the obsession. As they arrive in their vacation home, a huge, black storm cloud rolls in on the horizon

Great Movie, recommend watching it.

You mean like soccer, football, Tennis, Squash, or the myriad of other games played as sports?

This is excactly what some universities are starting to introduce. "Cumulative Thesis" is the key word. Sadly, this is not yet estbalished practice.

yet it is used to gatekeep positions. Contributions to science can be made without bundling them in a thesis, while being delivered to the whims of your supervisor.

The question is, do we really need a PhD reform? Or should we restructure the academic process to enable more people to pariticipate without going through the ardous process of writing a thesis? I have a PhD and work in an industrial research environment, with 50/50 PhD and non-PhD colleagues. There is no difference in quality of work and output. So maybe universities should be less discriminatory against non-PhD research fellows.

Because there is an inevitable translation step between "customer wants X" and "engineers have to build Y to achieve X". This takes time and is a non-trivial task. I'm an engineer myself and I'm glad there are people between me and the customers. They're there to shield engineers from unnecessary tasks and scope creep. They are the necessary abstraction layer between engineering and business.

The wikipedia article is pure gold!

The Bailey Ball was an Alpine Center attraction developed and tested, but never opened to the public, as a result of those tests. It consisted of a large foam sphere in which a rider could be secured, and then rolled downward. The plan was to do it on a track with PVC pipe as its outer rails, and one was built alongside a ski trail.[11]: 16:25

The designers neglected to take into account the tendency of PVC pipe to expand in heat. During the first test, with a state inspector present on a hot summer day, the ball, with a man inside testing it, went off the track as a result of the pipe expanding and bounded down the adjacent ski slope. It continued through the parking lot, across Route 94, and came to rest in a swamp. After it came to a natural stop at the bottom, the inspector left without saying anything and park management abandoned the project.[11]: 16:25 [43]

The Filmanalysis[0] - originally a German podcast, now also publishes episodes in english. While superficially a movie critic, the host Wolfgang M. Schmitt digs deep on the ideology side. He analyses the world view proposed by the creators of a piece, referencing works from sociology, psychology and philosophy.

[0] https://www.youtube.com/@thefilmanalysis2044

- be a better husband. I came off as selfish and inattentive in recent months. This needs to change.

- Spend more time with my family.

- Find a way to better engage with my town and community. I grew up where I live now, but still feel a little disconnected.

- Continue reading. Started reading again, this year. Want to keep up with that.

- Join the 1000lbs club. This is the follow up to this year’s goal of a 135 overhead / 225 Bench / 315 Squat / 405 DL

- finally finish my PhD. Still waiting for my Advisor to finish his report…

- relax more, stress less about income

I Think esp. fuelling stations along high-ways will improve their charging infrastructure. A lot of their revenue is non-food stuff like coffee and snacks. And longer wait times during charging could lead to more revenue there.

The wife and I started collecting cups and mugs whenever we travel. No touristy I<3 NYC, but if we run into a nice mug, we take it back home. Whenever you grab a coffee or tea, a little memory is there with it.

The general method presented here is Constraint Programming (CP), a sibling of SAT and ILP. In CP, you can formulate constraints over variables. Search is then performed branch-n-bound style. A variable is assigned, then constraints check if the assignment is legal and remove non-solutions from the other variable's domains.

If anyone is interested: An accessible python module for this is python-constraint. Gecode is a powerful, modern C++ CP Framework.

Maybe I falls out of line here, but I met the love of my live in 11th grade. We got married this September, at our 12th anniversary. There is no secret, just the will to commit to the relationship, mutual support and talking about issues when they arise.

I want to second this this method. It also works great in a research setting. There, the cool down also allows reflection on the past few week to update expectations and goals for next cycle. While not following this method strictly, I've been cycling through my work like this for past 12 months.

Maybe, but a real challenge of circuit layout and routing is that you want to generate (mostly) planar graphs, since your options for intersecting connection lines are limited and expensive. For the TSP, this does not really matter since the problem is more abstract.

No, they don't just random sample the base trees and hope for a better result. Their "actual work" is providing a 80 page proof for their improvement.

In the area of AutoTuning some machine learning methods are used. The goal is to optimize throughput or latency for compute intensive workloads. The machine learning model is trained with performance measurements on the hardware and then use it as proxy for real execution to try out different configurations.

Notable works:

AutoTVM - https://arxiv.org/abs/1805.08166

CHAMELEON - https://deepai.org/publication/chameleon-adaptive-code-optim...

GPU-Warp Scheduling - https://deepai.org/publication/rlws-a-reinforcement-learning...

I actually work in a corporate research lab, with 1000+ researchers working on all kinds of things. One thing that proves to be tricky is the "mission" these kinds of labs have. There is always a tension between direct work for business units on next years product, and research for novel ideas and methods. The former is what what keeps the lab afloat, financially. The latter is what is hoped to keep the company as a whole afloat in the future. But this includes investing money in projects that will not pan out and just burn money. Finding this balance is hard.

Tensions like these might very well be the reason corporate research has declined.

Try to find an activity you enjoy. You will never keep up a sports regime if you don't enjoy doin it. There are so many different effects in long term health, that the specific sports/exercise won't matter.

Also, consider this: Imagine you look back at your life, having spent countless hours on running, and never enjoying yourself.

I would strongly prefer some kind of hybrid model. To actually code or write something complex, WFH is great. But some interactions with colleagues just can't be replaced, even with slack and zoom.

The percentage should be flexible, depending on the current task and phase of the project.

Not directly with constraint programming itself. And also, I'm not doing a deep dive on these topics. But having an ideal of what is happening under the hood makes gives me more confidence in working with a tool and reasoning about choices.

I can second the experience that learning things in order to solve a problem works great. I struggle when I just have to sit down and lear something from scratch, without any direct application to the knowledge. But as soon I can project concepts and theories onto a problem I want to solve, I have no issue learning new things.

Currently I learn a lot about relational algebra and algebraic semirings, because they are part of the theory behind constraint programming, which I use in a current project.

This is just another test cardridge available for the existing on-site lab machine. The idea is that health-practitioners with access to the device can now test on premise, instead of sending samples to the lab. I can't image a lot local doctors have access to qPCR, where this device is useful for all kinds of other test.

Whats even more awesome: The ink is not erased, but literally made invisible by manipulating the molecues in the ink "responsible for the color". By reverting this chemical process, the ink can be made visible again!