At Google Cloud, we’re committed to making generative AI useful for everyone. Doing so requires more than making powerful foundation models available to businesses, governments, and developers. Models also need to be backed by platforms that make adoption faster and safer, with onramps to meet organizations wherever they are, regardless of their software or data science expertise. 

Today, we’re excited to announce general availability of Generative AI support on Vertex AI, giving customers access to our latest platform capabilities for building and powering custom generative AI applications. With this update, developers can access our text model powered by PaLM 2, Embeddings API for text, and other foundation models in Model Garden, as well as leverage user-friendly tools in Generative AI Studio for model tuning and deployment. Backed by enterprise-grade data governance, security, and safety features, Vertex AI can make it easier than ever for customers to access foundation models, customize them with their own data, and quickly build generative AI applications.

Vertex AI powers generative AI model customization for enterprise developers, data scientists, and everyone in between 

Foundation models are the starting point for creating customized generative AI applications—but models alone are not sufficient. That’s why in March, we announced Generative AI support on Vertex AI, the biggest-ever update to our machine learning platform, and began working with trusted testers. Now generally available to customers, Model Garden and Generative AI Studio leverage Google Cloud’s tight partnership with Google Research and Google DeepMind, making it easy for developers and data scientists to use, customize, and deploy models. 

Model Garden lets customers access and experiment with foundation models from Google and its partners, with over 60 models available and many more to come. In addition to making Model Garden, PaLM 2, and Embeddings API for text generally available, we’re also making our recently-announced Codey model for code completion, generation, and chat available for public preview. 

Along with these and other foundation models, Vertex AI offers a full ecosystem of tools to help builders tune, deploy, and govern models in production. For example, in May, we were the first enterprise ML platform to provide Reinforcement Learning with Human Feedback, or RLHF, which helps improve model usefulness and reduce cost. We’ve also upgraded Vertex AI’s suite of MLOps tools for model development and maintenance for customers who need to manage large models. With Generative AI Studio generally available, customers can now leverage an even wider range of tools, including multiple tuning methods for large models, that can significantly accelerate development of custom generative AI applications. 

We’re already seeing innovative results from early adopters via our trusted tester program and preview period. For example, leading global airline supplier GA Telesis is using our PaLM model on Vertex AI to build a data extraction solution that automatically synthesizes email orders and provides customers a quote. This eliminates the need for their sales teams to manually cross-reference emails with inventory availability. GitLab is leveraging our Codey model on Vertex AI for their “Explain this Vulnerability” feature, which gives their users a natural language description of code vulnerabilities, along with recommendations for resolving them. Canva, the visual communication platform, is using Google Cloud’s rich generative AI capabilities in language translation to better support its non-English speaking users, letting users easily translate presentations, posters, social media posts, and more into over a hundred languages. The company is also testing ways that Google’s PaLM technology can turn short video clips into longer, more compelling stories. 

And today, we’re pleased to share that Typeface, and DataStax are also building new generative AI capabilities with Vertex AI.    

Now is the time to build 

These announcements add to our news yesterday that we’ve added expanded access to Enterprise Search on Generative AI App Builder (Gen App Builder), allowing businesses to create custom chatbots and search engines that combine generative AI with Google’s semantic search technologies. Gen App Builder offers out-of-box solutions to common generative AI use cases, Vertex AI’s expansive platform capabilities can accelerate wide-ranging innovation, and growing ecosystem support from partners help our customers build freely. Together, these technologies and partnerships mean the full spectrum of developers and data scientists, from novices to seasoned experts, can build generative AI apps with enterprise-ready services on Google Cloud. 

As with our entire Cloud portfolio, Vertex AI and Gen App Builder help give customers complete control over their data; it doesn’t need to leave the customer’s tenant, is encrypted both in transit and at rest, and is not shared or used to train Google models. Google rigorously evaluates our new models to ensure they meet our Responsible AI Principles, and all of our generative AI offerings include the user security, data management, and access controls Google Cloud customers have come to expect. 

We’re grateful to our trusted testers for their integral role in bringing Generative AI support on Vertex AI to market, and we look forward to seeing what customers across all industries create with our growing catalog. To learn more about Google Cloud’s generative AI products, visit our solutions page, and to keep up with our latest AI news, don’t miss “The Prompt” or our generative AI primer for executives on Transform with Google Cloud.

You are what you eat. We’ve all been told this, but the truth is what we eat is often more complex than we are – genetically at least. Take a grain of rice. The plant that produces rice has 40,000 to 50,000 genes, double that of humans, yet we know far more about the composition of the human genome than of plant life. We need to close this knowledge gap quickly if we are to answer the urgent challenge of feeding 8 billion people, especially as food security around the globe is likely to worsen with climate change.

For this reason, AI company InstaDeep has teamed up with Google Cloud to train a large AI model with more than 20 billion parameters on a dataset of reference genomes for cereal crops and edible vegetables, using the latest generation of Google’s Tensor Processing Units (Cloud TPU v4), which is particularly suited for training efficiency at scale. Our aim is to improve food security and sustainable agriculture by creating a tool that can analyze and predict plants’ agronomic traits from genomic sequences. This will help identify which genes make some crops more nutritious, more efficient to grow, and more resilient and resistant to pests, disease and drought.

Genomic language models for sustainable agriculture

Ever since farming began, we have been, directly or indirectly, trying to breed better crops with higher yields, better resilience and, if we’re lucky, better taste too. For thousands of years, this was done by trial and error, growing crops year-on-year while trying to identify and retain only the most beneficial traits as they naturally arise from evolutionary mutations. Now that we have access to the genomic sequences of plants, we hope to directly identify beneficial genes and predict the effect of novel mutations.

However, the complexity of plant genomes often makes it difficult to identify which variants are beneficial. Revolutionary advances in machine learning (ML) can help to understand the link between DNA sequences and molecular phenotypes. This means we now have precise and cost-effective prediction methods to help us close the gap between genetic information and observable traits. These predictions can help identify functional variants and accelerate our understanding of which genes link to which traits – so we can make better crop selections.

Moreover, thanks to the vast library of available crop genetic sequences, training large models on hundreds of plant genomes means we can transfer the knowledge from thoroughly-studied species to those that are less understood but important for food production – especially in developing countries. And by doing this digitally, AI can quickly map and annotate the genomes of both common and rare crop variants.

One of the major limitations of traditional ML methods for plant genomics has been they mostly rely on supervised learning techniques. They need labeled data. Such data is scarce and expensive to collect, severely limiting these methods. Recent advances in natural language processing (NLP), such as Transformer architectures and BERT-style training (Bidirectional Encoder Representations from Transformers), allow scientists to train massive language models on raw text data to learn meaningful representations. This unsupervised learning technique changes the game. Once learned, the representations can be leveraged to solve complex regression or classification tasks – even when there is a lack of labeled data.

InstaDeep partners with Google Cloud to train the new generation of AI models for genomics on TPUs

Researchers have demonstrated that large language models can be especially effective in proteomics. To understand how this works, imagine reading amino acids as words and proteins as sentences. The treasure trove of raw genomics data – in sequence form – inspired InstaDeep and Google Cloud to apply similar technologies on nucleotides, this time reading them as words and chunks of genomes as sentences.

Moreover, the representations that the system learned improved in line with the size of the models and datasets, NLP research studies showed. This finding led InstaDeep researchers to train a set of increasingly larger language models on genomics datasets ranging from 1 billion to 20 billion parameters.

• Models of 1 billion and 5 billion parameters were trained on a dataset comprising the reference genomes for several edible plants, including fruit, cereal and vegetables for a total of 75 billion nucleotides.
• The training dataset must increase in the same proportion as the model capacity, recent work has shown. Thus, we created a larger dataset gathering all reference genomes available on the National Center for Biotechnology Information (NCBI) database including human, animal, non-edible plant and bacteria genomes. This dataset, which we used to train a 20-billion-parameter Transformer model, comprised 700 billion tokens, exceeding the size of most datasets typically used for NLP applications, such as the Common Crawl or Wikipedia dataset.
• Both teams announced that the 1 billion-parameter model will be shared with the scientific community to further accelerate plant genomics research.

The compact and meaningful representations of nucleotide sequences learned by these models can be used to tackle molecular phenotype prediction problems. To showcase their ability, we trained a model to predict the gene function and gene ontology (i.e. a gene’s attribute) for different edible plant species.

Early results have demonstrated that this model can predict these characteristics with high accuracy – encouraging us to look deeper at what these models can tell us. Based on these results, we decided to annotate the genomes of three plant species with considerable importance for many developing countries: cassava, sweet potato, and yam. We are working on making these annotations freely available to the scientific community and hope that these will be used to further guide and accelerate new genomic research.

Overcoming scaling challenges with massive models and datasets with Cloud TPUs

The compute requirement for training our 20 billion-parameter model with billions of tokens is massive. While modern accelerators offer impressive peak performance per chip, to utilize this performance often requires tightly coupled hardware and software optimizations. Moreover, maintaining this efficiency when scaling to hundreds of chips presents additional system design challenges. The Cloud TPU’s tightly-coupled hardware and software stack is especially well suited to such challenges. The Cloud TPU software stack is based on the XLA Compiler which offers out-of-the-box optimizations (such as compute and communication overlap) and an easy programming model for expressing parallelism.

We successfully trained our large models for genomics by leveraging Google Tensor Processing Units (TPUv4). Our code is implemented with the JAX framework. JAX provides a functional programming-based approach to express computations as functions that can be easily parallelized using JAX APIs powered by XLA. This helped us to scale from a single host (four chips) configuration to a multi-host configuration without having to tackle any of the system design challenges. The TPU’s cost-effective inter- and intra-communication capabilities led to an almost linear scaling between the number of chips and training time. This allowed us to train the models quickly and efficiently on a grid of 1024 TPUv4 cores (512 chips).

Conclusion

Ultimately, our hope is that the functional characterization of genomic variants predicted by deep learning models will be critical to the next era in agriculture, which will largely depend on genome editing and analysis. We envisage that novel approaches, such as in-silico mutagenesis – the assessment of all possible changes in a genomic region by a computer model – will be invaluable in prioritizing mutations that improve plant fitness and guiding crop improvements. Attempting similar work in wet-lab experiments would be difficult to scale and nearly impossible in nature. By making our current and future annotations available to the research community, we also hope to help democratize breeding technologies so that they can benefit all of global agriculture.

Further Reading

To learn more about the unique features of Cloud TPU v4 hardware and software stack we encourage readers to explore Cloud TPU v4 announcement. To learn more about scaling characteristics, please see this benchmark and finally we recommend reading PJIT Introduction to get started with JAX and SPMD parallelism on Cloud TPU.

This research was made possible thanks to the support of Google’s TPU Research Cloud (TRC) Program which enabled us to use the Cloud TPUv4 chips that were critical to this work.

NVIDIA GPU-powered instances on Google Cloud provide an optimal platform for organizations to develop their AI applications on the latest hardware and software stack, then seamlessly deploy those applications at scale in production.

Simplifying Workflows to Speedup AI Developments

NVIDIA recently announced the One Click Deploy feature on the NVIDIA NGC catalog, the hub for GPU-optimized AI software. Developed in collaboration with Google Cloud, this feature simplifies the deployment of AI software, to a single click from the NGC catalog.

This allows data scientists to deploy frameworks, software development kits and Jupyter Notebooks directly to Google Cloud’s Vertex AI Workbench, a new managed Jupyter Notebook service on top of Vertex AI, Google’s service for machine learning operations.

Under the hood, this feature launches the JupyterLab instance on Google Cloud Vertex AI Workbench with optimal instance configuration, preloads the software dependencies, and downloads the NGC notebook in one go.

NGC Catalog – One Stop for AI Software

NVIDIA is expanding the rich trove of NVIDIA AI software in the catalog to ensure AI practitioners have everything they need to get started — from frameworks to models.

All of the AI models in the catalog come with credentials. They’re like resumes that show the model’s skills, the dataset that trained it, how to use the model and how it’s expected to perform.

These model credentials provide transparency, which gives developers the confidence in picking the right model for their use case.

The NGC catalog also hosts Jupyter Notebooks tailored for the most popular AI/ML applications. Examples include:

Computer Vision – A collection of models for detecting human actions, gestures and more.

Automatic Speech Recognition – An end-to-end workflow for text-to-speech training.

Recommendation – A collection of example notebooks to help build end-to-end recommendation services.

Serve Robotics uses Vertex AI and the simple easy-to-use interface that the NGC One Click Deploy delivers.

“NGC catalog allows our ML research engineers to launch environments for experiments on Vertex AI with a single click. This saves us the efforts on ML infra setup and lets researchers focus on the ML problem in the computer vision and robotics space more efficiently.”—Kaiwen Yuan, Director of ML/Head of Perceptions & Predictions at Serve Robotics

Accelerate ML Deployments

Explore hundreds of Jupyter notebook examples for speech, computer vision and recommenders and, if you’re just getting started with AI, browse NVIDIA’s collection of Jupyter notebook examples and run it using the One Click Deploy feature on Google Cloud Vertex AI.

Every organization has its own unique data culture and capabilities. Yet each is expected to use technology trends and solutions in the same way as everyone else. Your organization may be built on years of legacy applications, you may have developed a considerable amount of expertise and knowledge, yet you may be asked to adopt a new approach based on a technology trend. On the other hand, you may be on the other side of the spectrum, a digitally native organization built with engineering principles from scratch without legacy systems but expected to follow the same principles as process driven, established organizations. The question is, should we treat these organizations in the same way when it comes to data processing? In this series of blogs and papers this is what we are exploring: how to set up an organization from the first principles from data analyst, data engineering and data science point of view. In reality, there is no such organization that is solely driven by one of these but it is likely to be a combination of multiple types. What type of organization you become is then driven by how much you are influenced by each of these principles. 

When you are considering what data processing technology encompasses, take a step back and make a strategic decision based on your key goals. This can be whether you optimize for performance, cost, reduction in operational overhead, increase in operational excellence, integration of new analytical and machine learning approaches. Or perhaps you’re looking to leverage existing employees’ skills while meeting all your data governance and regulatory requirements. We will be exploring these different themes and will focus on how they guide your decision-making process. You may be coming from technologies which are solving some of the past problems and some of the terminologies may be more familiar, however they don’t scale your capabilities. There is also the opportunity cost of prioritizing legacy and new issues that arise from a transformation effort, and as a result your new initiative can set you further behind on your core business while you play catch up to an ever changing technology landscape. 

Data value chain

The key for any ingestion and transformation tool is to extract data from a source and start acting on it. The ultimate goal is to reduce the complexity and increase the timeliness of the data. Without data, it is impossible to create a data driven organization and act on the insights. As a result, data needs to be transformed, enriched, joined with other data sources, and aggregated to make better decisions. In other words, insights on good timely data mean good decisions.

While deciding on the data ingestion pipeline, one of the best approaches is to look into the volume of data, the velocity of the data, and type of data that is arriving. Other considerations include the number of different data sources you are managing, whether you need to scale to thousands of sources using generic pipelines, whether you want to create one generic pipeline but then apply data quality rules and governance. ETL tools are ideal for this use case as generic pipelines can be written and then parameterized. 

On the other hand, consider the data source. Can the data be directly ingested without transforming and formatting the data? If the data does not need to be transformed and can be ingested directly into the data warehouse as a managed solution. This not only reduces the operational costs but also allows for more timely data delivery. If the data is coming in through an unstructured format such as XML or in a format such as EBCDIC and needs to be transformed and formatted, then a tool with ETL Capabilities can be used depending on the speed of the data arrival. 

It is also important to understand the speed and time of arrival of the data. Think about your SLAs and time durations/windows that are relevant for your data ingestion plans. This would not only drive the ingestion profiles but would also dictate which framework to use. As discussed above, velocity requirements would drive the decision-making process.

Type of Organization

Different organizations can be successful by employing different strategies based on the talent that they have. Just like in sports, each team plays with a different strategy with the ultimate goal of winning. 

Organizations often need to decide on what’s the best strategy to take in respect to data ingestion and processing – whether you need to hire an expensive group of data engineers, or exploit your data wizards and analysts to enrich and transform data that can be acted on, or whether it would be more realistic to train the current workforce to do more functional/high value work rather than to focus on building generally understood and available foundational pieces.

On the other hand, the transformation part of ETL pipelines as we know it, dictates where the load will be. All of these are made a reality in the cloud native world where data can be enriched, aggregated, and joined. Loading data into a powerful and modern data warehouse means that you can already join and enrich the data using ELT. Consequently, ETL isn’t really needed in its strict terms anymore if the data can be loaded directly into the data warehouse.

All of the above was not possible in the traditional, siloed, and static data warehouses and data ecosystems whereby systems would not talk to each other or there were capacity constraints in respect to both storing and processing the data in the expensive Data Warehouse. This is no longer the case in the BigQuery world as storage is now cheap and transformations are now much more capable without constraints of virtual appliances. 

If your organization is already heavily invested into an ETL tool, one option is to use them to load BigQuery and transform the data initially within the ETL tool. Once the as-is and to-be are verified to be matching, then with the improved knowledge and expertise one can start moving workloads into BigQuery SQL, and effectively do ELT. 

Furthermore, if your organization is coming from a more traditional data warehouse that extensively relies on stored procedures and scripting, then the question that one may ask is, do I continue leveraging these skills and expertise and use these capabilities that are also provided in BigQuery? ELT with BigQuery is more natural, similar to what’s already in Teradata BTEQ, Oracle PL/SQL but migrating from ETL to ELT requires changes. This change then enables exploiting streaming use cases, such as real-time use cases in retail. This is because there is no preceding step before data is loaded and made available.

Organizations can be broadly classified under 3 types as Data Analyst Driven, Data Engineering driven, and Blended organization. We will be covering a Data Science driven organization within the Blended category.   

Data Analyst Driven

Analysts understand the business and are used to using SQL/spreadsheets. Allowing them to do advanced analytics through interfaces that they are accustomed to enables scaling. As a result, easy to use ETL tooling to bring data quickly into the target system becomes a key driver. Ingesting data directly from a source or staging area then also becomes critical as it allows analysts to exploit their key skills using ELT and increases timeliness of the data. This is commonplace with traditional EDWs and realized by extended capabilities of using Stored Procedures and Scripting. Data is enriched, transformed, and cleansed using SQL and ETL tools act as the orchestration tools. 

The capabilities brought by cloud computing on separation of data and computation changes the face of the EDW as well. Rather than creating complex ingestion pipelines, the role of the ingestion becomes, bringing data close to the cloud, staging on a storage bucket or on a messaging system before being ingested into the cloud EDW. This then releases data analysts to focus on looking into data insights using tools and interfaces that they are accustomed to. 

Data Engineering / Data Science Driven 

Building complex data engineering pipelines is expensive but enables increased capabilities. This allows creating repeatable processes and scaling the number of sources. Once complemented with cloud it enables agile data processing methodologies. On the other hand, data science organizations allow carrying out experiments and producing applications that work for specific use cases but are not often productionised or generalized. 

Real-time analytics enables immediate responses and there are specific use cases where low latency anomaly detection applications are required to run. In other words, business requirements would be such that it has to be acted upon as the data arrives on the fly. Processing this type of data or application requires transformation done outside of the target.

All the above usually requires custom applications or state-of-the-art tooling which is achieved by organizations that excel with their engineering capabilities. In reality, there are very few organizations that can be truly engineering organizations. Many fall into what we call here as the blended organization.  

Blended org

The above classification can be used on tool selection for each project. For example, rather than choosing a single tool, choose the right tool for the right workload, because this would reduce operational cost, license cost and use the best of the tools available. Let the deciding factor be driven by business requirements: each business unit or team would know the applications they need to connect with to get valuable business insights. This coupled with the data maturity of the organization would be the key to making sure the right data processing tool would be the right fit. 

In reality, you are likely to be somewhere on a spectrum. Digital native organizations are likely to be closer to being engineering driven, due to their culture and business that they are in. However, brick and mortar organizations would be closer to being analyst driven due to the significant number of legacy systems and processes they possess. These organizations are either considering or working toward digital transformation with an aspiration of having a data engineering / software engineering culture like Google. 

The blended organization with strong skills around data engineering, would have built the platform and built frameworks, to increase reusable patterns would increase productivity and then reduce costs. Data engineers focus on running Spark on Kubernetes whereas infrastructure engineers focus on container work. This in turn provides unparalleled capabilities as application developers focus on the data pipelines and even the underlying technologies or platforms changes code stays the same. As a result, security issues, latency requirements, cost demands and portability are addressed at multiple layers. 

Conclusion – What type of organization are you?

Often an organization’s infrastructure is not flexible enough to react to a fast changing technological landscape. Whether you are part of an organization which is engineering driven or analyst driven, organizations frequently look at technical requirements that inform which architecture to implement. But a key, and frequently overlooked, component needed to truly become a data-driven organization is the impact of the architecture on your data users. When you take into account the responsibilities, skill sets, and trust of your data users, you can create the right data platform to meet the needs of your IT department as well as your business.

To become a truly data-driven organization, the first step is to design and implement an analytics data platform that meets your technical and business needs. The reality is that each organization is different and has a different culture, different skills, and capabilities. Key is to leverage its strengths to stay competitive while adopting new technologies when it is needed and as it fits to your organization. 

To learn more about the elements of how to build an analytics data platform depending on the organization you are, read our paper here.

Verve Group is an ecosystem of demand and supply technologies fusing data, media, and technology to deliver results and growth to both advertisers and publishers – no matter the screen or location, no matter who, what, or where a customer is. Classifying massive amounts of this unstructured data at scale is the first step in helping to surface relevant, high-quality content to users—and that’s where natural language processing (NLP) comes in.

Verve Group uses the NLP API from Google Cloud’s Vertex AI to fetch data for their internal content classification quality verification and as an additional source for building categorization models. By leveraging the NLP API’s Content Classification models, which are now generally available and offer Google’s latest large language model (LLM) technology, Verve Group powers classification through an updated and expanded training data set with over 1,000 labels and support for 11 languages (Chinese, French, German, Italian, Japanese, Korean, Portuguese, Russia, Spanish, and Dutch join previously-available English). 

Verve Group has been using Google Cloud’s NLP API since day one, because of both the ease of implementation and the quality compared to competing NLP products. With documentation that is “comprehensive and self-explanatory,” the NLP API “allows for fast adoption and implementation, from test models all the way to production,” said Rami Alanko, GM of Verve Group.

Leveraging the NLP API has facilitated Verve Group’s fast go-to-market motions by enabling its customers to quickly discover and classify new content. “Operating on a global level with tens of different regions and languages, we have still been able to maintain high quality for our product and high retention rates with our clients,” Rami shared. “In a recent client case, we achieved 82% improvement in CTR when optimized with content quality measurements enabled by the API. In another client case, we drove brand safety risk down to 0.16% from 4% thanks to classification quality. Along with the new functionalities of the Google NLP, I can only see this trend continuing to strengthen.”

Verve Group is excited to further expand their NLP use cases by leveraging the new Content Classification models, which have already helped them expand their classification inventory, improve the quality and performance of their quality verification for customers, and unlock new use cases for NLP. “Our classification model accuracy improved 41% using Google NLP as a verification partner,” said Rami.  

Additionally, Verve Group is now using the API for metadata analysis on a  large image database. “We browse the database and run the image metadata via our classification. This flow enables us to classify images reliably aligned with our standard classification.  We pretty much use the same data flow for our runtime in-app textual content analysis, therefore allowing for close to real-time consumer engagement,” Rami added.

To learn more about how companies are leveraging NLP API from Google Cloud Vertex AI, click here, and to learn more about Google Cloud’s work with foundation models and generative AI, read The Prompt on Transform with Google Cloud.