Every week brings tons of news, opinions, and discoveries in the world of machine learning and AI. But with our Weekly Roundup, you can easily catch up with the most important information and learn what’s happened in the industry.
Here are a few best articles from the last week. Enjoy the read!
Weekly Roundup: August 18-24
» 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. 🙂
» Too many AI researchers think real-world problems are not relevant by Hannah Kerner on MIT Technology Review | August 18
One might think that applying machine learning to real-world problems is one of the main goals of ML in general. But it turns out that it’s not always the case. This article is an important opinion, definitely worth keeping in mind.
» Chatbots Are Machine Learning Their Way To Human Language by Adrian Bridgwater on Forbes | August 20
Is machine learning going to reduce the language barrier between people and AI? It seems researchers are making huge progress in this field, and chatbots are learning more and more about understanding our dynamic language.
» Machine learning reveals role of culture in shaping meanings of words by Rachel Nuwer on Phys | August 17
Even though you can translate one word to different languages, it turns out that the meaning of it in various cultures can be slightly different. If you’re interested in cultural differences and its effect on language, have a look at this piece. It’s really fascinating how the place we live in and the tradition we grow up in can change our understanding of words!
» Can artificial intelligence prompt a creative revolution? by Sam Fletcher on Tech HQ | August 20
It’s always been underlined that our creativity is one of the characteristics that distinguish us from the AI, that machines can’t really recreate it. But who says they can’t help us enhance it and stimulate new ideas?
» Is Artificial Intelligence (AI) medicine racially biased? by Rod McCullom on Genetic Literacy Project | August 24
As you can see, the problem of racial inequality can appear in all fields of life and science. In the article, we can read that the extent of racial bias in the medical AI is unknown, but noticing it is the first step to addressing the issue. Very interesting read, that shows how much there’s still to explore and describe in this field.
👉 The story was originally published on Undark.
» AI may not predict the next pandemic, but big data and machine learning can fight this one by Anna Solana on ZDNet | August 21
The news about pandemic has taken over most of the information channels in the last few months. It’s been really overwhelming and sometimes difficult to follow. Now, that it becomes calmer, it’s worth looking at the whole situation with perspective and searching for possible solutions. Read about how AI may help with that.
» 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! I hope you found something of interest in this weekly roundup. Don’t forget to check our blog for more inspiring articles.
👉 Came across an interesting ML article? Or maybe you wrote one yourself and would like to share it with other people? Let us know, we’ll spread the news in our weekly roundup!
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 ->