Neural nets trained on sensitive data, like medical data or social security numbers, can “accidentally” memorize it, leaving it vulnerable to hackers.
“We hope to raise awareness that it’s important to consider protecting users’ sensitive data as machine learning models are trained. Machine learning or deep learning models could be remembering sensitive data if no special care is taken.”
What is a blockchain?
Blockchain is a distributed ledger, a system that is maintained on a peer-to-peer network and uses cryptography to secure transactions.
There is a lot of discussions the past years about the revolution predictive analytics will bring to the businesses. But, predictive analytics is not something new. It used to be a part of statistics! Corporations with the resources were doing it for decades. Nowadays, however, it is more accessible, due to technological advancements in the areas of computing power, software and storage, the abundance of big datasets, and advances in algorithm research. Any student can now use the cloud to perform an experiment that a few years ago would have been possible to be performed only by large corporations or government agencies.
The challenge is how can we apply advanced analytics, and AI in an enterprise environment, efficiently and effectively.
The first time I saw predictive analytics live, in praxis, to provide useful information was in the mid-90s. As a student and aspiring engineer, I was enrolled in a Total Quality/Six Sigma class. During a field trip we visited a Japanese run, car manufacturing plant where, to my astonishment, blue-collar workers were applying regression in order to predict when to change tools!
As a consultant with its almost 20 years’ experience of providing high-level professional services to large corporations and global players in the manufacturing, financial, media, and telecommunications industry, I am committed upon providing value to our clients, and metadata management is an excellent way of doing so.
Let’s see what metadata is for your business: