There is real progress being made in AI! But, with all the hype, it is difficult to distinguish reality from fiction. This curated newsletter will help you distill the real AI from the noise — stay informed with how AI is being used today to automate well-defined tasks, advancements in platforms/tools to apply AI, and progress in research in defining the art of possible! A weekly newsletter for your Sunday morning reading!
- AI robo-advisors in Finance: AI is helping human financial advisors by relieving them from having to perform mundane portfolio monitoring and administrative tasks that currently take a significant portion of their time. When allocations fall outside of certain parameters for specific clients, an AI-based system can trigger it into the monitor of the human advisor. “Analysis of vast quantities of historical and financial data will uncover alpha opportunities that traditional analysis would otherwise overlook and give robo-advisors an edge over passive strategies and AI can process big data swiftly, allowing robo-advisors to adapt to changing market conditions and consumer behaviors much quicker in order to make better investment decisions.” Betterment uses AI to reduce taxes on transactions where machine learning algorithms select the specific tax consequences of the transactions. SigFig also uses its AI engine automatically to allocate assets and determines which investments will result in minimum taxes.
- Zillow using AI to predict floor plan dimensions: Zillow 3D Home app allows uploading a series of photographs of the house to stitch into a 3D tour. Using AI, it then generates a floor plan from the uploaded photos and predicts the dimensions and square footage of rooms based on the photos (accuracy of the predicted dimensions is under testing).
- Siemens using AI for performance optimization of a Turbine Control: By learning the turbine’s control strategy, AI outperforms manually tuned turbines in terms of the NOx emissions generated. A picture is worth 1000 words:
- Google open-sourced Model Search, a platform that helps researchers develop the best ML models, efficiently and automatically. Instead of focusing on a specific domain, Model Search is domain agnostic, flexible, and is capable of finding the appropriate architecture that best fits a given dataset and problem, while minimizing coding time, effort, and compute resources. It is built on Tensorflow and can run either on a single machine or in a distributed setting.
- AWS announced SageMaker Pipelines that automates different steps of the ML workflow, including data loading, data transformation, training and tuning, and deployment. Amazon SageMaker Pipelines offers DevOps best practices of Continuous Integration and Continuous Delivery (CI/CD) applied to machine learning (known as MLOps) to automate and scale ML model building and deployment pipelines.
- Microsoft Research open-sourced ResNet-50, a pre-trained model built using the Bing search engine’s web-scale image data. The model achieves interesting performance across seven computer vision benchmarks.
Happy reading! Follow on medium to get notified for the next weekly newsletter!