Dear Cognitive RoundUp Readers,
Every week SwissCognitive - The Global AI Hub - selects 9 articles from the Artificial Intelligence Universe to share with our community.
Among some other topics, this week’s articles focus on how Machine Learning, IoT, and with that, Artificial Intelligence can help the world's poorest, how we can build AI that we can rely on, and in the era of AI what lessons we can learn from the dinosaurs.
We wish you happy reading and a great Sunday!
The SwissCognitive Team
“We’re using IoT and ML to optimise water spend, placement of pesticides, dealing with weeds and other pests, so food production costs are dropping. These are the things that proportionally impact the poorest people in the world.”
As AI technologies standardize across industries, becoming an AI-fueled organization will likely be table stakes for survival. That means rethinking the way humans and machines interact.
Recent air travel accidents raise critical two questions: when should we trust fully-automated computer systems with autonomous decision making and how do we allow humans to gain control if things go wrong?
Applying AI has shown to result in tremendous business growth. But without the right artificial intelligence approach, it won’t be possible to achieve the success we aim for. How does one get their AI approach right and how do we efficiently build an AI use case?
AI has a data quality problem. In a survey of 179 data scientists, over half identified addressing issues related to data quality as the biggest bottleneck in successful AI projects.
The ancestors of modern birds were the sole survivors of one of the most severe mass extinction events in the history of the world. In the fourth industrial revolution era, enterprises and human workers are also at risks. Being proactive and being able to adapt are going to be key for survival.
Even if we humans aren’t always good at explaining our choices, at least we can try to explain them somehow. A deep learning system can’t do this yet, and therefore, identifying which input data the systems are triggering on to make decisions is the direction we need to head.
A new model developed by researchers creates richer, more easily computable representations of how individual amino acids determine a protein’s function, which could improve machine-learning tasks in protein design, drug testing, and other applications.
The most important contribution of blockchain technology in AI is to make data available to all the stakeholders without any kind of centralized and corporate control. This means all the benefits and advantages of AI can be shared democratically.