InfoQ Homepage Google Content on InfoQ
-
Google Uses AutoML to Discover More Efficient AI Training Algorithm
Researchers at Google have open-sourced EvoLved sIgn mOmeNtum (Lion), an optimization algorithm for training neural networks, which was discovered using an automated machine learning (AutoML) evolutionary algorithm. Models trained with Lion can achieve better accuracy on several benchmarks than models trained with other optimizers, while requiring fewer compute cycles to converge.
-
Google AI Updates Universal Speech Model to Scale Automatic Speech Recognition beyond 100 Languages
Google AI has recently unveiled a new update for their Universal Speech Model (USM), to support the 1,000 Languages Initiative. The new model performs better than OpenAI Whisper for all segments of automation speech recognition.
-
Google Service Weaver Enables Coding as a Monolith and Deploying as Microservices
Google has released Service Weaver, an open-source framework for building and deploying distributed applications. The Go-based framework includes a set of programming libraries that enable writing applications as a single modular binary. The other component is a set of deployers that allow for configuring the runtime topology and running the application locally or in the cloud.
-
Android 14 Brings Partial Support for OpenJDK 17, Improved Privacy and Security, and More
Google has announced the first Android 14 beta, which provides support for over 300 OpenJDK 17 classes. Additionally, it implements a number of features aimed to keep malicious apps at bay and extends support for foldable form factors, battery usage optimization, and more.
-
Google Adds New Pricing Model to Its Security Command Center
Google recently announced several new updates to its Security Command Center (SCC) with a pay-as-you-go pricing model and two capabilities: deployments at the project level and self-service activation.
-
Go 1.20 Improves Compiler Performance and Extends Coverage Support
The latest release of the Go language, Go 1.20, improves compiler performance, bringing it back in line with Go 1.17. Additionally, the language now supports conversion from slice to arrays and revises struct comparison.
-
Google Unveils MusicLM, an AI That Can Generate Music from Text Prompts
Google researchers have introduced MusicLM, an AI model that can generate high-fidelity music from text. MusicLM creates music at a constant 24 kHz throughout a number of minutes by modeling the conditional music generating process as a hierarchical sequence-to-sequence modeling problem.
-
Google Delivers Comprehensive Cloud Infrastructure Reliability Guide
Google recently delivered a cloud infrastructure reliability guide combining best practices and expertise from its engineers for its customers.
-
Log Analytics Feature in Cloud Logging Now Generally Available
Google recently made its Cloud Logging Log Analytics feature generally available (GA), allowing users to search, aggregate, and transform all log data types, including application, network, and audit logs.
-
Google Storage Transfer Service Now Supports Serverless Real-Time Replication Capability
Recently Google announced the preview support for event-driven transfer capability for its Storage Transfer Service (STS), which allows users to move data from AWS S3 to Cloud Storage and copy data between multiple Cloud Storage buckets.
-
Google AI Unveils Muse, a New Text-to-Image Transformer Model
Google AI released a research paper about Muse, a new Text-To-Image Generation via Masked Generative Transformers that can produce photos of a high quality comparable to those produced by rival models like the DALL-E 2 and Imagen at a rate that is far faster.
-
Android Extension SDK Aims to Simplify the Use of Modular System Components
Introduced with Android 10, Modular System Components enable updating end-user devices outside of the normal Android release cycles. The new Extension SDK framework, now public, aims to make their integration simpler for developers.
-
Google Publishes Technique for AI Language Model Self-Improvement
Researchers at Google and University of Illinois at Urbana-Champaign (UIUC) have published a technique called Language Model Self-Improved (LMSI), which fine-tunes a large language model (LLM) on a dataset generated by that same model. Using LMSI, the researchers improved the performance of the LLM on six benchmarks and set new state-of-the-art accuracy records on four of them.
-
Google Releases Open-Source Vulnerability Scanning Tool
Google has released OSV-Scanner, an open-source front-end interface to the Open Source Vulnerability (OSV) database. The OSV database is a distributed, open-source database that stores vulnerability information in the OSV format. The OSV-Scanner assesses a project's dependencies against the OSV database showing all vulnerabilities relating to the project.
-
Google's Code-as-Policies Lets Robots Write Their Own Code
Researchers from Google's Robotics team have open-sourced Code-as-Policies (CaP), a robot control method that uses a large language model (LLM) to generate robot-control code that achieves a user-specified goal. CaP uses a hierarchical prompting technique for code generation that outperforms previous methods on the HumanEval code-generation benchmark.