Discover Microsoft’s “Bonsai Brain”: a low-code AI platform that accelerates the development of AI-based automation

Microsoft’s recent ongoing project called Bonsai Brain is dedicated to modeling and creating a low-code AI component that can be applied to various standalone tasks and applications. The Bonsai brain has been trained and practiced to handle unforeseen scenarios and maintain operations. Its key selling feature is the significant decrease in downtime resulting from improved production efficiency. Larger neural networks must be developed for automation tasks, but the Bonsai brain works without trained or emulated neural networks. Users can create their own bespoke AI models using the Bonsai Brain Interface and implement them appropriately without the need for additional resources.

To simulate and train the Bonsai brain for all unpredictable conditions and to ensure that wiser autonomic systems are developed, the Bonsai brain platform relies heavily on deep reinforcement learning concepts. Three guiding principles govern the functioning of the Bonsai brain platform, with the Integrate component serving as the central principle. This part is responsible for merging Bonsai brain training simulations with real circumstances and giving the training process proper feedback. Through the contribution of the Integrate component, the second component, known as Train, is in charge of training and modeling the brain. The final export component of the platform is a fully trained and simulated Bonsai Brain that will be made accessible as a Linux container installed on-premises or in the Azure environment. Two control conditions must be met to train the Bonsai brain into the platform. To ensure that Bonsai Brain is reliable and works as expected, the first criterion checks that the precision of each simulated action must be exact. The second requirement ensures that the probability of reversing an erroneous action taken by the brain must be high or fast.

Five essential elements form the basis of the entire bonsai brain. The agent of the Bonsai platform that will be trained and simulated to achieve the necessary goals is called the Brain. The second element of the Bonsai platform is the simulator, which simulates the brain to enable learning from various situations. The observations will be the input to the simulator, and the output of the simulator will be the different sets of actions that the Bonsai Platform Bonsai Brain will perform. The workspace is one of the parts of the Bonsai platform, which houses all the brains and simulators developed on the platform. An element of the Bonsai platform is iteration, which trains the brain to perform a particular action for each set of simulations. The platform’s brain therefore considers each action as an iteration. The last part of the Bonsai platform, called Episode, is used to establish a cut-off point for iterations of the platform. The model’s ability to mimic and train the brain in accordance with industry standards and subject matter expertise ensures that the brain simulation retains its robustness. This is one of its defining characteristics. It can be modeled to quickly adapt to immediate production changes required.

With the help of the Bonsai mastermind, Microsoft hopes to remove unnecessary code and implement efficient and reliable AI models. The Bonsai Brain uses deep reinforcement learning methods to create efficient AI models that quickly simulate and build reliable AI models. According to the researchers, once the Bonsai brain is completed, it will be an essential component of many automation systems and will be integrated into many AI models.

References:

  • Documentation: https://docs.microsoft.com/en-us/bonsai/
  • https://docs.microsoft.com/en-us/bonsai/product/
  • https://docs.microsoft.com/en-us/bonsai/product/components/simulation


Khushboo Gupta is an intern consultant at MarktechPost. She is currently pursuing her B.Tech from Indian Institute of Technology (IIT), Goa. She is passionate about the fields of machine learning, natural language processing and web development. She likes to learn more about the technical field by participating in several challenges.


Comments are closed.