Tag AI

Posts: 20

Microsoft the AI Powerhouse

I've been a long term shareholder of MSFT and I've been rewarded quite well. Under the leadership of CEO Nadella, Microsoft has become and AI powerhouse, and I believe he'll win the cloud computing wars. Right now Amazon is the dominant player but based on what I see in Azure's development, I think both companies will be 'neck to neck' in a few short years.

Investors have rewarded the stock with new highs and crossing (before retracing just below) a $1 Trillion dollar valuation. Sure Apple did it first and then Amazon, but the one who wins is the one with staying power.

MSFT AI Powerhouse

This one key thing caught my eye from a recent Bloomberg article. The change in perception from a 'fixed mindset' to a 'growth mindset.'

Microsoft marketers like to attribute its reemergence as a tech power to a sort of cultural rehab, involving what Nadella calls corporate “empathy” and a shift of his team from a “fixed mindset” to a “growth mindset.” The reality of the company’s turnaround was more painful, according to interviews with more than four dozen current and former executives, board members, customers, and competitors. Under Nadella, it cut funding to Windows and built an enormous cloud computing business—with about $34 billion in revenue over the past year—putting it ahead of Google and making progress in key areas against the dominant player, Amazon Web Services. “I don’t know of any other software company in the history of technology that fell onto hard times and has recovered so well,” says Reed Hastings, CEO of Netflix Inc. via Bloomberg

I know how hard it is to change a massive juggernaut like Microsoft from the inside. Ballmer certainly didn't do it but I think Nadella will. The numbers won't lie and I expect to add more to my position over time.

comments

Making AI Happen Without Getting Fired

I watched Mike Gualtieri's keynote presentation from H2O World San Francisco (2019) and found it to be very insightful from a non-technical MBA type of way. The gist of the presentation is to really look at all the business connections to doing data science. It's not just about the problem at hand but rather setting yourself up for success, and as he puts it, not getting fired!

My notes from the video are below (emphasis mine):

  • Set the proper expectations
  • There is a difference between Pure AI and Pragmatic AI
  • Pure AI is like what you see in movies (i.e. ExMachina)
  • Pragmatic AI is machine learning. Highly specialized in one thing but does it really well
  • Chose more than one use case
  • The use case you choose could fail. Choose many different kinds
  • Drop the ones that don't work and optimize the ones that do
  • Ask for comprehensive data access
  • Data will be in silos
  • Get faster with AutoML
  • Data Scientists aren't expensive, they need better tools to be more efficient
  • Three segments of ML tools
    • Multimodel (drag and drop like RapidMiner/KNIME)
    • Notebook-based (like Jupyter Notebook)
    • Automation-focused (like Driverless AI)
  • Use them to augment your work, go faster
  • Warning: Data-savvy users can use these tools to build ML. Can be dangerous but they can vet use cases
  • Know when to quit
  • Sometimes the use case won't work. There is no signal in the data and you must quit
  • Stop wasting time
  • Keep production models fresh
  • When code is written, it's written the same way and runs the same forever
  • ML Models decay, so you need to figure out how to do it at scale
  • Model staging, A/B testing, Monitoring
  • Model deployment via collaboration with DevOps
  • Get Business and IT engaged early
  • They have meetings with business and IT, get ducks in a row
  • Ask yourself, how is it going to be deployed and how it will impact business process
  • Ignore the model to protect the jewels
  • You don't have to do what the model tells you to do (i.e False Positives, etc)
  • Knowledge Engineering: AI and Humans working together
  • Explainability is important

comments

Weaponizing AI

File this under "no shit Sherlock," but hackers are already weaponizing machine learning.

The AI, named SNAP_R, sent simulated spear-phishing tweets to over 800 users at a rate of 6.75 tweets per minute, luring 275 victims. By contrast, Forbes staff writer Thomas Fox-Brewster, who participated in the experiment, was only able to pump out 1.075 tweets a minute, making just 129 attempts and luring in just 49 users. via Gizmodo

In reality the above example is just a bunch of loops and stuff, it's what the tweets contain and to whom it's sent to that makes all the difference. That 'intelligence' is probably generated from some machine learned model or "AI".

Artificial intelligence, and machine learning in particular, are perfect tools to be using on their end.” These tools, he says, can make decisions about what to attack, who to attack, when to attack, and so on.

Yes, propensity to buy to click models and even some NLP will get people to infect their machines.

Chalk this up to another abuse, yet innovative way, to use machine learning.

comments

Millennials can't catch a break

This is just nuts. Millennials just can't seem to catch a break. Now AI is coming for their jobs.

Research released by Gallup on Thursday indicates a collision between technology and “business as usual” is coming soon, and the fallout will be ugly, especially for Millennials. Automation and artificial intelligence (AI) are among the most disruptive forces descending upon the workplace, says the Gallup report, and 37% of Millennials “are at high risk of having their job replaced by automation, compared with 32% of those in the two older generations.”[via Forbes]

So how can they stay relevant? Look for new trends in hiring. The top one I can think of is Data Science.

If you’re considering a career move, get a beat on what jobs are trending up (software engineer) and which ones are on their way out (reporter). You can boost your skills through a boot camp or with a traditional degree, no matter what your industry is, but know that some companies may prefer a regular degree over a boot-camp certificate or DIY learning.

But those industries might be susceptible to offshoring.

Though the Bureau of Labor Statistics (BLS) says that programmer and coder jobs will decline 8% due to outsourcing to other countries from 2014 to 2024, there will still be plenty of work, and in many cases, it will be too unwieldily to move massive operations overseas.

So in other words, Millennials can't seem to catch a break. If I were part of that creative and awesome generation, I'd probably go the route of entrepreneurship.

comments

Labeling Training Data Correctly

When you’re dealing with a classification problem in machine learning, good labeled data is crucial. The more time you spend labeling training data correctly, the better. This is because your model’s performance and deployment will depend on it. Always remember that garbage in means garbage out.

Thoughts on labeling data

I recently listened to a great O’Reilly podcast on this subject. They interviewed Lukas Biewald, Chief Data Scientist and Founder of CrowdFlower. CrowdFlower provides their clients with top notch labeled training data for various machine learning tasks, and they’re busy!

The few bits that caught my ear were how much of the training data is used in deep learning. They’re also seeing more image labeled data for self driving cars.

The best part of the interview as Lukas’s discussion on using a Raspberry Pi with Tensor Flow! How cool is that?

The Podcast

https://soundcloud.com/oreilly-radar/data-preparation-in-the-age-of-deep-learning?in=oreilly-radar/sets/the-oreilly-data-show-podcast


Originally published at Neural Market Trends.

comments