Welcome to Nural's newsletter focusing on how AI is being used to tackle global grand challenges.
Packed inside we have
- Uber Eats and Nuro sign a 10-year deal to do robot food delivery in California and Texas
- Crypto interlude: Ethereum upcoming transition from proof-of-work to proof-of-stake will slash energy consumption by 99%
- and A.I. Is Making It Easier Than Ever for Students to Cheat
If you would like to support our continued work from £1 then click here!
This weeks newsletter is brought to you by Wynter.
Wynter, a customer feedback tool, is looking for data science professionals (CEOs, senior leaders, MLEs, data scientists) to join its paid research panels.
Participate in messaging research panel surveys, get paid for your insights.
It takes ~1-15 min to be part of a single research panel. Super low time commitment.
Join Wynter’s Paid Data Science Panels | Learn More
Key Recent Developments
Uber Eats and Nuro sign a 10-year deal to do robot food delivery in California and Texas
What: Uber customers may soon have their orders delivered by a self-driving delivery pod after the company signed a 10-year deal with autonomous driving start-up, Nuro.
The deal comes only two years after Uber sold off their own self driving unit to Aurora. The partnership with Nuro will mean that from Autumn, Mountain View, California and Houston, Texas will have Uber driverless pods on the streets.
Key Takeaway: Self-driving vehicles are very much the present, so it's important that our thinking around its safety have accelerated at the same rate as the technology. For now, limitations such as speed limits and restricting their activity to certain cities have been put in place.
Crypto interlude: Ethereum upcoming transition from proof-of-work to proof-of-stake will slash energy consumption by 99%
What: The Ethereum merge is estimated to occur over the next day and this represents a significant change in the operating mechanism for the blockchain underpinning the world's second largest crypto token, Ether.
The merge indicates the point at which the Ethereum chain changes its consensus mechanism from “proof of work” to “proof of stake.” Alongside other improvements. The change should enable increase scalability, security and speed.
Key Takeaways: Outside of the interesting effects that this will have for the economic incentives of the chain, the merge also brings a significant environment upgrade. Energy usage is predicted to drop 99% following the transition to a less energy intensive proof of stake approach.
Analysing Employee Attrition in Healthcare Data and Predicting Outcomes
What: Healthcare is an industry where its employees are known for working long, arduous hours and this was only made worse throughout and following Covid-19. In response to the high global attrition rate, a PhD student has taken an ML driven approach to analysing employee attrition rates in healthcare.
Key Takeaway: This is a great example of the value of datasets produced by large orgs. This was a dataset produced by IBM which is s publicly free to use, modify and share under the creative commons license. While none of the factors (long hours, low pay, and low supply in the workforce) found to be driving high turnover are particularly surprising, this is a great introduction to the data science workflow.
AI Ethics & 4 Good
Other interesting reads
Cool companies found this week
Midjourney - Midjourney is an independent research lab exploring new mediums of thought and expanding the imaginative powers of the human species.
They are a small self-funded team focused on design, human infrastructure, and AI.
Albedo - High resolution, aerial quailty images from space
DeepMind's efforts to create an AI system that can navigate obstacles
AI/ML must knows
Foundation Models - any model trained on broad data at scale that can be fine-tuned to a wide range of downstream tasks. Examples include BERT and GPT-3. (See also Transfer Learning)
Few shot learning - Supervised learning using only a small dataset to master the task.
Transfer Learning - Reusing parts or all of a model designed for one task on a new task with the aim of reducing training time and improving performance.
Generative adversarial network - Generative models that create new data instances that resemble your training data. They can be used to generate fake images.
Deep Learning - Deep learning is a form of machine learning based on artificial neural networks.
Nural Research Founder
If this has been interesting, share it with a friend who will find it equally valuable. If you are not already a subscriber, then subscribe here.
If you are enjoying this content and would like to support the work financially then you can amend your plan here from £1/month!