We are happy to announce MultipleXLab: a modular, mobile, and cost-effective plant imaging setup. The system can continuously monitor thousands of seeds from germination to root development.
Paper: https://t.co/C1WWXGiXdb
Our latest research has just been published in @SciReports.
We developed a deep neural network to help dermatologists identify cells in Tzanck smears.
https://t.co/lOPLrnfVTB
#MachineLearning#ArtificialIntelligence
In essence, #MachineLearning is all about finding the signal inside the data where:
Data = Signal + Noise
If you are building a cat classifier these would be:
Data: Cat images
Signal: "Catness"
Noise: Background, light etc.
Natalia kindly invited me to talk about possible career paths for Ph.D.s outside academia.
We discussed why I did an industrial Ph.D. (@dindustrialscat) and how I ended up at @ipsumio doing machine learning.
https://t.co/EqHshBTGoJ
How to predict probabilities for each cell independently?
Several people have asked this regarding our recent study:
https://t.co/4oB5hiESNR
For each output node, there is a sigmoid activation function that gives a number between 0 and 1, which is the predicted probability.
We are 1 year old already.
While we celebrate it, we would like thank all our business partners for this amazing first year.
Here is what we did so far:
https://t.co/8GtB7CFYeF
Physics and engineering rely heavily on mathematical models, developed using scientific laws. Usually, as the model accuracy increases, their complexity increases as well.
#AI vs. #humans: Enemies or friends? If you agree that, we need another point of view, read our founder @malicannoyan's thoughts on #AI for human intelligence: https://t.co/TAP5iaDRGH
#MachineLearning
Tips for training #NeuralNetworks. These are heuristics rather than laws. You can start building a NN using these and then adjust it according to the problem you would like to solve. This is not an exhaustive list so please feel free to add more below.
An ultra-compact #particle#size#analyser using a CMOS image sensor and #MachineLearning at a fraction of the cost and size of traditional particle size analyzers. Check out our new paper to learn more: https://t.co/EJJTUKXfTj