Machine learning is fascinating. New things happen every second while we’re busy performing our daily tasks. If you want to know what big things have happened over the last two weeks, don’t go anywhere.
Here are the best picks from the last two weeks from the world of the machine learning. Enjoy the read!
Weekly Roundup: November 3 – 16
» Neptune.ai blog – as always, make sure to visit our blog to find out interesting and in-depth articles on machine learning from the last week.
We’ve published a lot of new content recently so don’t wait, go through our blog!
» AI pioneer Geoff Hinton: “Deep learning is going to be able to do everything” by Karen Hao on MIT Technology Review | November 3
An interesting and insightful (and short) interview with professor Geoff Hinton about AI. Make sure to check it out!
» Why 90 percent of all machine learning models never make it into production by Rhea Moutafis on Towards Data Science | November
Quite a concise piece on the difficulties with AI projects and why they fail. Mostly because Companies are lacking leadership support, effective communication between teams, and accessible data.
It’s also a good guide on how to deliver a successful AI project and what to focus on.
» CSIRO to use artificial intelligence, machine learning, and sensors to end plastic waste by Aimee Chanthadavong and Asha Barbaschow on ZDNet | November 9
We’re facing a huge climate change and environmental problems but there may be a way to prevent it. The Commonwealth Scientific and Industrial Research Organisation (CSIRO) has announced partnerships with Microsoft, Hobart City Council, and Chemistry Australia to address Australia’s plastics waste issue.
» This could lead to the next big breakthrough in common sense AI by Karen Hao on MIT Technology Review | November 6
Researchers are teaching giant language models how to “see” to help them understand the world. What are the results and where will it lead? Read the story to find out!
» The 600-Year-Old Vatican Library Is Using Artificial Intelligence to Ward Off Hackers Targeting Its Digital Collections by Brian Boucher on artnet news | November 9
The Vatican Library is digitizing its collection of more than 80,000 manuscripts using the same artificial-intelligence technology as big tech companies like eBay, T-Mobile, and Samsung.
» When Machine Learning Knows Too Much About You by Eric Siegel on KDnuggets | November 15
Machine learning is great and brigs lots of good to our world. But there are limits. What are they? What do machines know about you and is it safe? Learn more in this smart article! 🤖
» Training Facial Recognition on Some New Furry Friends: Bears Lesley Evans Ogden on The New York Times | November 11
Ed Miller and Mary Nguyen are Silicon Valley software developers help keep track of individual bears with machine learning algorithms. It can help not only in recognizing different bears but in saving them, and can later be used for other animals. 🐻
» KDnuggets™ News of the week with top stories and tweets of the past week, plus opinions, tutorials, events, webinars, meetings, and jobs.
» Old but gold, the reliable Reddit thread on ML for more news on machine learning. There’s always something for everyone – tips, tricks, hacks, and more news.
That’s all folks! Have you found something of interest in this weekly roundup? We hope you got inspired! Don’t forget to check our blog for more inspiring articles.
And if you came across an interesting ML article, or maybe wrote one yourself and would like to share it with other people, let us know, we’ll spread the news in our weekly roundup! See you next week!
ML Experiment Tracking: What It Is, Why It Matters, and How to Implement It
Jakub Czakon | Posted November 26, 2020
Let me share a story that I’ve heard too many times.
”… We were developing an ML model with my team, we ran a lot of experiments and got promising results…
…unfortunately, we couldn’t tell exactly what performed best because we forgot to save some model parameters and dataset versions…
…after a few weeks, we weren’t even sure what we have actually tried and we needed to re-run pretty much everything”
– unfortunate ML researcher.
And the truth is, when you develop ML models you will run a lot of experiments.
Those experiments may:
- use different models and model hyperparameters
- use different training or evaluation data,
- run different code (including this small change that you wanted to test quickly)
- run the same code in a different environment (not knowing which PyTorch or Tensorflow version was installed)
And as a result, they can produce completely different evaluation metrics.
Keeping track of all that information can very quickly become really hard. Especially if you want to organize and compare those experiments and feel confident that you know which setup produced the best result.
This is where ML experiment tracking comes in.Continue reading ->