HNHacker News
TopNewBestAskShowJobs

aabajian

1,743 karma · joined December 29, 2010

aabajian aaaaaat gmail.com 697b79
submissionscomments
aabajian··on Show HN: AI Grant – Get $2,500 for your AI project in minutes
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 have users fill out an application without logging in and then showing them a survey that they may not have time to complete.
aabajian··on Machine Learning Crash Course
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 individual images were only 375 GB, but took up 1.5T when converted to Numpy matrices. Why? The arrays were converted to standard int arrays (32-bit x 3 channels) when they could've fit into short (8-bit x 3 channels). Each image was 4x as large as it needed to be.

You can certainly use high-level ML tools (like Keras), but it takes a great deal of work to wrangle your data into a usable format, and even more knowledge to debug an ineffective network.

aabajian··on The Problem of Doctors’ Salaries
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 highest earners in any country, it's the price you pay to ensure the brightest students are attracted to medicine despite the long training. I do agree with the following:

A second route would be to end the requirement that foreign doctors complete a U.S. residency program in order to practice medicine in the United States. This means setting up arrangements through which qualified foreign doctors could be licensed to practice in the United States after completing an equivalent residency program in another country. The admission of many more doctors would put downward pressure on the pay of doctors in the United States, as insurers would have a new pool of physicians to add to their networks who will accept somewhat lower compensation.

I've met many foreign-graduates who have already completed residency. In makes ZERO sense to have these individuals repeat a residency in the U.S. They have already undergone 12+ years of training, actively treat patients in their home countries, and there's a need for physicians in the U.S. This is very much cartel-esque of the AMA. I know it would result in lower salaries, but this would (hopefully) translate into lower-cost medical school in the U.S.

This is a situation that may benefit from tech solution. There are already certain states allowing foreign medical graduates to practice in the U.S. If you're looking for a "startup pain point", let me present you with the two sided market:

1. There were ~7,500 international medical graduates who applied to residency in the U.S. (http://www.nrmp.org/wp-content/uploads/2016/09/Charting-Outc...) Presumably, these are individuals who want to practice in the U.S., but must undergo a U.S. residency. I don't know how many of them have already completed a residency in their home country, but let's conservatively say 25% or ~1,800.

2. The AAMC is estimating a physician shortage of 30,000 by 2030 (https://news.aamc.org/medical-education/article/new-aamc-res...).

Even if all of the foreign train medical graduates were allowed to practice in the U.S. w/o residency, it still wouldn't be enough. Maybe there's a startup opportunity that matches foreign MDs to positions in the U.S. As I mentioned, states like Missouri have already allowed some non-residency-trained physicians to practice (https://www.statnews.com/2017/05/15/missouri-doctor-dearth/). My preference would be to have a foreign doctor who has undergone residency in his or her home country rather than a U.S. grad with no residency.

aabajian··on TensorFlow Mask R-CNN code for pixelwise object detection and segmentation
Wow, this is fascinating. From a radiology perspective, it could be the missing method for segmenting findings inside a convoluted radiograph.
aabajian··on Bay Area hammered by loss of jobs: Lack of affordable housing strangles hiring
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 buy purchasing Sun Microsystem's campus right next to the bridge.

Also, note that SF isn't the fastest growing housing market anymore, it's Seattle (https://www.seattletimes.com/business/real-estate/seattle-no...). Seattle didn't feel as much housing-pressure with Microsoft due to the remote location of its campus in Redmond/Bellevue. That's changing with Amazon; the sales giant is located in prime real estate, just like the tech companies on the peninsula.

aabajian··on Ask HN: What is your yearly passive income?
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 "sms reminders." I did a ton of work, made an iPhone app, changed the billing system four times and rebuilt the interface three times.

I now have ~60 paying customers, but it only makes $300 per month after all expenses. I don't think I've recouped the man-hours spent, but I've certainly learned a lot along the way. I learned Python/Django (website back-end), Objective-C and Swift (iOS app), and lots of fun stuff about timezones and calendaring APIs.

aabajian··on Gluon – Deep Learning API from AWS and Microsoft
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.
aabajian··on Bubble – Visual Programming
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 latter - premature optimization - that is a pain in lower-level languages like C++, and which stops you from getting the correct result.
aabajian··on Bubble – Visual Programming
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 LabView function, a simple coefficient of variation - mean divided by standard deviation, in C++. LabView can call C++ DLLs. The resulting function was 5x faster and meant that our imaging of blood flow could be done in real-time vs. recorded and post-processed.

FWIW, it took 10x as long to get the DLL call to work as it did to drag-and-drop the coefficient of variation box into the GUI. There are certainly rapid prototyping benefits to be had with higher-level languages. Nowadays I prototype in Python and rewrite line-by-line in C++ if I really need the speed. I think GUI programming languages might serve a similar purpose.

aabajian··on Ask HN: What's the next big advance in AI you're looking forward to most?
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 solution would be some combination of video or audio that records the clinical encounter and automatically generates a note based on what was discussed and performed. It falls under Paul Graham's "Schlep Task" (http://www.paulgraham.com/schlep.html); you'd have to work with individual clinicians, get their (and the patient's!) approval to record the encounter, record it from multiple angles, build tech into dumb devices (e.g. stethoscope, Dopplers for pulses, O2 saturations, and somehow use machine learning to integrate all that sensor data into a plain-text note.

It's probably the #1 problem from a provider's viewpoint right now, especially on the primary care side. Any individual provider will see 10-20 patients per day and write just as many notes.

Edit: I'd add that you'd also have to have access to existing notes because progress notes typically summarize the patient's prior visits.

aabajian··on Wal-Mart Offers Linux as Windows Alternative (2003)
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.
aabajian··on Introducing Pytorch for fast.ai
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 is the engineer who uses Autodesk to design car parts: He/she may not need to know the detailed implementation of the 3D graphics to use the software effectively.

The difference is that Autodesk relies on a mature, deterministic technology (3D graphics rendering). Deep learning is a stochastic process that depends on the data and the model. The training code, and especially the framework hooks, is the least important part. The example they give is three lines of code to train a cat vs. dog classifier. I've tried this classifier on a different binary image classification task: livers with and without tumors. It didn't work very well. There's lots of reasons: little variability between images, grey-scale images, different resolutions, etc. You can tweak the network, throw in more middle layers, try different kinds of layers, whatever, to get better results. All of that is guesswork if you don't understand what the CNN is doing at each stage. At this point in time you do need a formal education in linear algebra, calculus and statistics to investigate why a model does/does not work. It's not enough to know how to use the libraries.

On the flipside, you also need to know how to manipulate data and parse it into the correct format. This generally requires a year or two of programming practice in a good scripting language like Python. I will echo their thoughts that Ian Goodfellow's Deep Learning Book is remarkably lacking in this area. As a simple example, you cannot even use AlexNet without pre-processing your images to be 227x227 or 224x224 for GoogleNet. That's 10,000 images resized, labeled and loaded into the model before training can take place.

tl;dr IMHO in terms of being a competent user of deep learning: mathematics >= programming >>> knowing how to use a framework

aabajian··on Where will artificial general intelligence come from?
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 computing, you're still bottlenecked by the final integration step (two cores sharing the result of their calculation). There is no central place in our brains where thoughts begin and end. A CPU's clock and ALU are simple not analogous to the human mind. As far as we know, human intelligence is a constant, dynamic interaction between all neurons in our brains, any one of which is capable of originating a signal. I personally think we will develop AGI, but with a different computational model. I don't know enough about quantum computing to even comment on it, but I do have a background in medicine (MD) and computer science (MSCS).
aabajian··on Real estate site Redfin files for IPO
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.
aabajian··on 3D printers start to build factories of the future
I had an ah-ha moment reading about digital light synthesis. This is how 3D printing should be done; not just printing molten materials, but chemically forming them. It might be possible to print CPUs this way.
aabajian··on Is a $70 Hackintosh Any Good?
I think the takeaway (along with the lucky discount from the university), is that the consumer needs for CPU/GPU power has not increased that dramatically in the last five years. VR, 3D gaming, deep learning, 4K rendering, etc. are all edge-cases. The MacBook Pro 13" is likely designed without these uses in mind.
aabajian··on The Dark Side of Doctoring
It's worth noting that the incredible success of Epic EMR software is because it tightly controls all of the administrative billing issues, NOT because it makes clinician's lives easier. There are endless check boxes in Epic and each site has its own interface. It's a huge mess and difficult to navigate...but arguably still better than the other vendors.
aabajian··on What was it like to self-learn programming before Stack Overflow? (2016)
I grew up in the mountains of Southern California. No mentors, nothing other than an 80386 running DOS 5.0 and QBASIC. There was a book on BASIC at the local library which became a staple of my programming education.
aabajian··on Ask HN: What are the most important problems in your industry?
Don't have time right now to go into detail, but in radiology the biggest problem is rising volumes with lower reimbursement. I addressed this in another thread, but it's an arms race between vendors to maximize radiologist throughput via tools such as dictaphones, templates, computer-aided diagnoses, and now, machine learning.
aabajian··on Deep Patient Predicts Patients Future from Health Records
This is quite interesting. The obvious business case is for missed-diagnoses reimbursement. U.S. hospital reimbursement is based on a monstrous coding system (e.g. ICD-10). A subset of these codes are Hierarchical Condition Categories (HCCs). Hospitals can bill based on the constellation of HCCs that apply to a patient (e.g. morbid obesity, rheumatoid arthritis, COPD, etc.). Their system could identify all HCCs for each patient and submit missed HCCs for reimbursement.

Edit: This is not a new idea, there are lots of companies already doing this using natural language processing.

aabajian··on San Fran is so expensive, some tech companies can't convince employees to move
I almost purchased a condo in East Palo Alto back in 2013 for $300K. That same 1-bedroom is now selling for $680K; biggest missed opportunity of my life. Here it is on Zillow:

https://www.zillow.com/homes/for_sale/East-Palo-Alto-CA/cond...

There are no 1 bedrooms from Santa Clara to Burlingame for < $350K, and only one two-bedroom for less than $500K.

aabajian··on How much money doctors actually make
Computer-science wise: I'm a rather good developer, according to my supervisors. I'm quite comfortable with Java, Python and a bit of Swift. I love writing code, starting with BASIC when I was 11 years old.

Medicine-wise: I'm in the bottom quartile of my class. I'm terrible at sitting there and studying. It's a completely different mind set. I did really poorly on UMSLE Step 2 in particular; most med students do +20 points higher relative to Step 1...I did 40 points lower. That test is entirely memorization, we're talking memorizing immunization schedules for children, the screening guidelines for low-dose CT, the warning signs of an acute abdomen, and so on...

In terms of benefits, I'm much faster at research data analysis. I'm doing imaging work and I can take a DICOM, convert it to NIfTI, run an algorithm in Python or MATLAB, and dump the results in under and hour. That kind of stuff is second-nature to me, but takes my fellow lab mates an indeterminate amount of time. I've written a number of 20-line scripts for my classmates that do basic stuff like vertical lookups in Excel and searching for files on the filesystem. Really simple stuff that would take a person a long time by hand.

One more thing, the MCAT was a blast for me. It was a wonderful test. It emphasizes integration and application of what you've learned on novel problems, very much like computer science. The USMLEs were hell. I got 98th percentile on the MCAT; 40th percentile on the USMLE.

aabajian··on How much money doctors actually make
I actually got my master's in the middle of medical school (I took a leave of absence). My undergrad degree is in computer science as well. It's my long-term goal to start a radiology informatics company.
aabajian··on How much money doctors actually make
I'm a graduating medical student with a master's in computer science. In a lot of ways CS is the polar opposite of medicine; one values derivation and automation while the other emphasizes memorization and one-on-one interactions.

Surprisingly, teamwork is one area where CS wins out. I've found that despite the professed importance of the healthcare team, healthcare often doesn't function like a team at all. Doctors, nurses, PAs, patient care assistants, administrative staff, can make or break your life by providing support or creating obstacles. I think it's a result of a inherently tiered system with a single person (e.g. the attending) on top who decides the treatment plan. In CS you almost never have a single person who is overly-trained relative to peers. Maybe you'll have a better developer, UI designer or manager, but none of these attributes makes that person the sole decider. The pay structure in medicine reflects its hierarchical system: software managers don't earn 10x what software developer make, whereas attendings easily make 10x a nurse's salary.

aabajian··on Unsupervised sentiment neuron
My point would be to see how much the model is relying on punctuation. It could provide insight as to why character-based models outperform word-based models for sentiment analysis.
aabajian··on Unsupervised sentiment neuron
I'm trying to understand this statement:

"The sentiment neuron within our model can classify reviews as negative or positive, even though the model is trained only to predict the next character in the text."

If you look closely at the colorized paragraph in their paper/website, you can see that the major sentiment jumps (e.g. from green to light-green and from light-orangish to red) occur with period characters. Perhaps the insight is that periods delineate the boundary of sentiment. For example:

I like this movie. I liked this movie, but not that much. I initially hated the movie, but ended up loving it.

The period tells the model that the thought has ended.

My question for the team: How well does the model perform if you remove periods?

aabajian··on Verizon is mashing Yahoo and AOL into a new company called Oath
This is interesting. At first I was on the side of Nissan Computer Corp., but on thinking about it: it's only because of Google that I know Nissan.com is not the motor company. This must have been a major pain for Nissan motors before Google. Nissan.com probably has many, many more hits due to visitors trying to find Nissan Motors. In such cases, I think we need to have a means to acknowledge that there is probably a stronger argument in favor of Nissan motors owning the domain.
aabajian··on Jeff Bezos Is Now the World's Second Richest Person
Worth noting that Gates leads by a $10 billion margin.
aabajian··on Ask HN: Hacker News users reading habits
I'm surprised there are no comments on your question...so let me be the first. I'm a medical student about to start residency. I also have a master's degree in computer science.

I read a lot of articles about AI in medicine, pretty much anything I can get my hands on. I also read generic tech articles related to everything from Nintendo Switch, Tesla, Brain-Computer interfaces, and other popular media articles.

-How many articles do you read each day? Likely 10+. These aren't high-brow articles, just random blog posts and pop culture tech. I read about 2-3 research abstracts per day in medicine and maybe skim the text of 1-2 articles.

-They're usually related to your job or to some side projects? Usually they are related to my interest in medicine or technology. Sometimes they are related to my job (I work as a part-time developer / data scientist). I also run a small website (https://www.cronote.com). I encountered a number of issues with time-zone switching and the daylight savings change on March 12th. Read about 20 articles having to do with correctly implementing timezones in Python.

-Do you usually read about a variety of topics or it's focused in 2 or 3 topics only? Topics cover a vast span of medicine and computer science. I enjoy computer science more than medicine so it's a 20:80 split.

-Do you usually read during some time of the day or it's usually random? I read whenever I'm behind my computer, usually alternating between work and browsing the Internet. This amounts to ~5 hours per day.

aabajian··on Distill: a modern machine learning journal
I already have a nomination. The guy who wrote this blog post:

http://adilmoujahid.com/posts/2016/06/introduction-deep-lear...

It's the only way I could get a working model of Caffe while understanding the data preparation steps. I've already retrofitted it to classify tumors.

← PreviousPage 6 of 10Next →