InfoQ Homepage Data Science Content on InfoQ
-
Microsoft Announces General Availability of Jupyter Notebooks Support for Cosmos DB
Recently Microsoft announced the general availability of Jupyter notebooks support for Cosmos DB, providing integrated support for running queries directly against all data models. As these notebooks run directly inside Cosmos DB; this allows for analyzing and visualizing the data directly from the Azure portal, without the need to extract the data.
-
The First AI to Beat Pros in 6-Player Poker, Developed by Facebook and Carnegie Mellon
Facebook AI Research’s Noam Brown and Carnegie Mellon’s professor Tuomas Sandholm recently announced Pluribus, the first Artificial Intelligence program able to beat humans in 6 player hold-em poker. In the past years, computers have progressively improved, beating humans in checkers, chess, Go, and the Jeopardy TV show. Poker poses more challenges around information asymmetry and bluffing.
-
Google Adds New Integrations for the What-If Tool on Their Cloud AI Platform
In a recent blog post, Google announced a new integration of the What-If tool, allowing data scientists to analyse models on their AI Platform – a code-based data science development environment. Customers can now use the What-If tool for their XGBoost and Scikit Learn models deployed on the AI Platform.
-
Databricks MLflow Integration Now Generally Available
Databricks recently made MLflow integration with Databrick notebooks generally available for its data engineering and higher subscription tiers. The integration combines the features of MLflow with those of Databrick notebooks and jobs. MLflow provides the following three main capabilities: experiment tracking, projects, and MLflow models.
-
Google Releases Google-Landmarks-V2, a Large-Scale Dataset for Landmark Recognition & Retrieval
Google has released Google-Landmarks-v2, an improved dataset for Landmark Recognition & Retrieval, along with Detect-to-Retrieve, a Tensorflow codebase for large-scale instance-level image recognition. Two companion Kaggle competitions based on Google-Landmarks-v2 were also launched. With over 200,000 landmarks in 5 million images, it is the largest landmark dataset ever published.