Artificial Intelligence and Machine Learning allows us to develop powerful applications with unprecedented benefits for science, cyber security, healthcare, manufacturing, utilities, enterprise, government, telecoms, and practically for every field in our lives, as they combine human intelligence and machine compute power. The big challenge is to proactivity identify when AI/ML applications deviate from normal behavior and avoid undesired results to mission critical and business applications. Take for example an AI/ML video surveillance application that monitors suspicious travelers in international airports. If the application misses out suspicious travelers in Chicago O’hare International Airport, you will assume that the AI/ML model requires a fix or at least some tuning. This requires huge technical efforts, time, and data scientist resources that may not be available. However, the root cause of the problem may not be related to the AI/ML model and you may be looking in the wrong direction. The common AI/ML monitoring solutions that analyze the AI/ML models, would not be helpful in such cases. If, on the other hand, you would perform a holistic monitoring process, where you compare the software version, the operating systems, the operation of the sensors that feed the AI/ML application, in this case the video surveillance cameras, and the lighting between thousands of video surveillance deployments across dozens of airports, you may find out that the root cause of the problem may be related to a new and mal-performing video surveillance cameras uniquely deployed in the Chicago O’Hare airport. It is the ‘IoT crowd wisdom’, or in this case, the wisdom gained from inspecting thousands of video surveillance cameras, that would allow you to identify the root cause of the problem. The only way to quickly drill down to the root cause of the problem is by correlating all the parameters impacting the application performance, starting with the code, the sensors and the environment. Now, wouldn’t you like your AI/ML application to benefit from such monitoring capabilities? Well, you can. Yokaranda is holistic, SaaS based, AI/ML monitoring solution that collects parameters impacting the application performance and identifies deviation from normal behaviour. The R&D team just needs to include the Yokaranda open source library in a process that takes no more than 10 minutes, and you are done. Once you do that, you have added IoT crowd wisdom to your AI/ML application, making it more robust and reliable.
Comentarios