aabajian

15 ms·

121.
by aabajian 8y ago
Amphetamines were popularly used to suppress appetite in the mid-20th century, but addiction and abuse proved disastrous. He glosses over amphetamines. They are highly effective at weight loss. So much so that ADHD patients have the ste
122.
by aabajian 8y ago
LOL, logged in at work just to make this same comment. You never know the love of a car until you've owned one.
123.
by aabajian 8y ago
I'm a radiology resident with a master's in computer science. It's great that healthcare organizations are finally releasing larger datasets. There's much progress to be made in medical ML/AI. However, my attitude t
124.
by aabajian 8y ago
This is a fascinating paper. It demonstrates how rooted deep learning is in classical mathematics, and how libraries like TensorFlow/Keras/PyTorch/etc. have both helped and hurt its progress. A software developer can build
125.
by aabajian 8y ago
I agree with the author overall, but I think she picks a poor example for dictation. “The bareheaded man needed a hat” falls squarely in the domain of contextually-aware models and would likely be transcribed correctly using deep learning
126.
by aabajian 8y ago
I have an S2000. It only ever came in manual. It is insanely fun to drive.
127.
by aabajian 8y ago
wwww.cronote.com Simple website for sending reminders. It didn't make any money for 4 years until I added commonly-requested features (templates, recurring reminders, and variables for recipient names). Now I have about 60 paying custo
128.
by aabajian 8y ago
I did my master's degree in computer science at Stanford, but now I'm a radiology resident at the University of Washington in Seattle.
129.
by aabajian 8y ago
I did a prelim-surgery intern year. The hours are worse than medicine, but there are way less "mandatory" activities (no lectures, no endless rounding, etc.). For the most part, you're on the floor waiting for pages. I used t
130.
by aabajian 8y ago
This is amazing. I graduated from medical school two years ago with $300,000 in debt. I'm a software developer on the side and would love to stop residency to try a healthcare startup and/or pursue a research, but the risk is too
131.
by aabajian 8y ago
Wow, I just spent several hours today trying to integrate Google's text-to-speech with a simple webapp. Here's a comparison. Google: "There is a multiloculated fluid collection adjacent to the liver." AssemblyAI: "T
132.
by aabajian 8y ago
Not to get existential here, but the analogy to human neural networks is compelling. The algorithm efficiently redesigns itself to fit the data, just like neural synapses reconnect one another as knowledge is learned, to optimize the storag
133.
by aabajian 8y ago
I think something that's overlooked is the importance of parents in admissions. Case in point: Jewish culture values education in the same way that Asian families value education. As a result, a large number of white students admitte
134.
by aabajian 9y ago
Pimsleur audio books all the way. It's spaced-learning/repetition with conversations.
135.
by aabajian 9y ago
Article opens with, "There are lots of reasons to avoid processed foods. They’re often packed with sugar, fat and salt, and they tend to lack certain nutrients critical to health, like fiber." Diet coke has no sugar, no fat, and l
136.
by aabajian 9y ago
Does Tesla record when people override autopilot? Seems like this would be a good source of information about sections of road with obvious problems.
137.
by aabajian 9y ago
I completed the whole application and then it presented the quiz/survey. I didn't have time to complete it so I closed the window. I never created an account...does your application save automatically? Seems like poor design to ha
138.
by aabajian 9y ago
While I like the idea, in principle, that you don't need a CS education to use AI/ML, I doubt it. Here's a problem that cropped up today: Our instance ran out of hard drive space on a training set of ~400,000 images. The indi
139.
by aabajian 9y ago
I'm a surgical intern right now, it's not fun. I'll be doing radiology next year, and very much looking forward to it. I'm biased, but I don't agree with the premise of the article; Doctors should be among the hig
140.
by aabajian 9y ago
Wow, this is fascinating. From a radiology perspective, it could be the missing method for segmenting findings inside a convoluted radiograph.
141.
by aabajian 9y ago
I place blame on the location of the tech hubs in the Bay Area. Not one mega-company, save Tesla, has setup shop in East Bay. There is affordable housing across the water, and new construction as well. Facebook has mitigated this somewhat
142.
by aabajian 9y ago
I run a side project for sending business reminders: https://www.cronote.com Cronote made next to no money for 5 years, but slowly migrated its way to the top of Google for search of "text message reminders" and "
143.
by aabajian 9y ago
As a long-time Java developer, Python was a beauty. It brings back the joy of programming and makes data manipulation a breeze. C# is better than Java, but it's still not as elegant/simple/clean as Python for data science.
144.
by aabajian 9y ago
You use black-box unit tests developed against the Python code. You then write identical code in C++ that produces the same output (i.e. passes the unit tests). Then you apply optimizations/refactoring specific to C++. It is the latt
145.
by aabajian 9y ago
LabView is the classic example of graphical programming used in professional practice. It works well, but has a large learning curve. My first paid programming job was an undergraduate summer research project. I was tasked with re-writing a
146.
by aabajian 9y ago
I'm a medical resident. I see a patient in clinic, do a history and physical, sit down at a computer, and write a note describing what I just did. That's medical documentation, and it's a burdensome, growing problem. The so
147.
by aabajian 9y ago
I bought SuSE Linux at Best Buy 15 years ago. This was before I had a fast-enough internet connection to download it. I remember there were a few other distros available as well.
148.
by aabajian 9y ago
I get the motivation behind fast.ai's lessons, but I question whether it's the right approach. Their goal is to make deep learning accessible to programmers by reducing the mathematical background required. An analogous situation
149.
by aabajian 9y ago
These arguments about AGI all seem to overlook that our computational model is still very Turing-constrained. It's a clock-based, sequential model where each calculation is taken linearly in time. Even with multi-core and distributed c
150.
by aabajian 9y ago
To get rid of real estate agents: You have to find a way to create a free multiple listing service (satellite imagery maybe, looking for FOR SALE signs?). It's the MLS monopoly that has kept the real estate industry in the dark ages.

More ›