Elemental Cognition (EC) is a leading artificial intelligence company that develops scalable and transparent AI solutions for businesses and organizations, helping them with tasks like pharmaceutical research and complex travel planning. EC’s years of research and innovation have led to significant progress in AI that leverages large language models (LLMs) while simultaneously prioritizing correctness and transparency. Their AI platform combines the latest generative AI technology from Vertex AI with EC’s proprietary reasoning and deep natural language understanding technology to produce solutions that can understand complex problems, collaborate with users to solve them, and explain the results.
EC’s AI platform is delivered as SaaS software deployed on Google Cloud, leveraging key capabilities, including Google Kubernetes Engine (GKE) and BigQuery, to achieve the scale, speed, and ease of management necessary to support customers with big data and complex problems.
Building on its AI platform, EC currently provides solutions for research and knowledge discovery and complex problem solving. EC’s research and knowledge discovery product, Cora, enables faster, higher-quality research for mission-critical discovery and evidence-based decisions. EC’s expert problem-solving product, Cogent, tackles the most logically intricate and complex scheduling, configuration, planning, and optimization problems quickly and with provable correctness.
Both products leverage the strengths of large language models available in Vertex AI and apply them in combination with EC’s structured knowledge and logical reasoning technology to generate trustworthy, evidence-based results. The following figure provides a high-level view of the EC AI platform and how EC’s solutions build on the platform to support use cases across industries, from pharmaceutical research to complex travel planning.
Use-cases: Challenges and Solutions
EC’s products leverage Google Cloud AI to enable powerful solutions to challenging business use cases, such as those described below.
Expert Problem Solving: Travel
Taking a round-the-world trip can be a lifetime highlight, but planning and booking one is often tedious: mixing and matching airfares and availability from different airlines, optimizing travel distances and layover times, and trying different travel dates to find the best prices. Anyone who has done this knows how complicated it can be to get everything right.
EC’s customer Oneworld offers competitively priced round-the-world tickets to numerous destinations globally. However, these itineraries require following complex fare rules. Prior to EC’s solution, customers would often struggle to book online, most often requiring assistance from a travel agent, which increases costs and deters customers from booking.
Other digital automation solutions struggle with this use case. Workflow-based chatbots cannot handle the massive complexity of this booking problem: over 10^34 possible itineraries make a flowchart untenable to build and maintain. LLMs alone are not up to the task of the precise logical reasoning required to solve complex planning reliably and correctly.
EC, on the other hand, can deliver fluent and reliable solutions to this kind of problem.
Elemental Cognition and Oneworld have created Journey, an AI agent that helps customers plan and book complex round-the-world itineraries online. Journey helps customers navigate complex fare rules and ever-changing flight availability, intelligently helping them understand and manage tradeoffs, all while satisfying personal customer preferences.
The Journey agent is built on EC’s Cogent platform (shown in the figure above), using GKE and Vertex AI LLMs. Cogent interacts with business analysts in natural language to capture Oneworld’s rules, constraints, and policies in a readable document. Cogent turns these business documents into dynamic, multimodal applications that help customers plan, book, and buy efficiently.
Cogent made the Journey agent possible. Journey quadrupled conversion rates, enabling OneWorld customers to book online quickly without human assistance. It resulted in happy customers, more purchases, and reduced costs.
Research and Discovery: Drug Discovery for BioPharma
A biopharma company can spend over $2 billion to take a drug from initial discovery to approved use in the market. Identifying high-quality targets at the beginning of this process is essential for operating as efficiently as possible while generating the highest quality drug target leads. Comprehensive and fast literature review at the earliest stages is a key ingredient of efficient and effective target identification.
Key sources for this literature review include, among others, peer-reviewed research articles available through PubMed, information on clinical trials and outcomes, patents related to the disease and potential drug targets, and NIH grant awards. The challenge is effectively searching through all this content to find relevant information quickly. Moreover, doing this often requires connecting multiple bits of information described across different documents and sources.
To meet the requirements of pre-clinical research and discovery for biopharma companies, EC created Cora for Life Sciences. By combining the wide coverage for natural language understanding provided by the BERT and T5 word embedding models, the generative AI capabilities of PaLM 2, and EC’s proprietary semantic analysis and reasoning capabilities, Cora automatically analyzes and ingests content in the life sciences domain.
The figure above shows the high-level architecture and workflow. Data flows into the Cora content analysis and ingestion process, where Cora automatically extracts rich knowledge structures using word embeddings, semantic parsing, and deep automatic analysis of concepts, relations, and qualifiers. Cora automatically identifies key concepts in the domain, such as genes, proteins, biomarkers, symptoms, etc., and how these concepts relate to each other. After additional processing to cluster, type, and link concepts, Cora loads the analysis results into the knowledge index.
At runtime, the Cora semantic query engine and logical reasoning engine leverage the knowledge index and any domain models to process the query, analysis, dialog, and evidence summarization requests from the front-end API. Cora uses PaLM 2 to both interpret natural language questions and generate provably correct search result summaries grounded in the specific evidence found by Cora. Any GUI can be connected to the Cora APIs, though Cora also includes a default UI/UX with the SaaS product.
The base system has already analyzed and ingested all of the content in PubMed (the portion freely available for commercial use), and is immediately available for SaaS use in Google Cloud. Cora can also easily ingest proprietary customer content and satisfy all of the customer’s security and privacy requirements, leveraging Google Cloud’s comprehensive data management and security support.
The overall system provides a powerful research and discovery platform for researchers conducting preclinical drug discovery literature research. In one evaluation, Cora reduced the time required to perform a drug repurposing research task from two weeks to less than two hours. Cora’s usefulness extends well beyond pre-clinical literature research, however, and can be applied to content analysis across the entire drug discovery life cycle.
Considerations and Tradeoffs
The first consideration with any application of generative AI is understanding the requirements for veracity. For applications requiring creativity, or where secondary validation of results is expected or required, applying generative AI is relatively straightforward. EC’s primary focus, however, is on applications where veracity, transparency, and provable correctness are essential. To provide trustworthy and accurate results from LLMs, EC applies them with constraints and guardrails that ensure the results are grounded in reliable evidence and logically correct explanations. EC solutions cannot produce hallucinations because they never ask the LLM to generate answers without grounding them in a human-approved domain model or reliable evidence.
Another consideration is the cost and speed of invoking an LLM. This is where the Vertex AI solutions excel and provide better response time at lower cost than competitors. There are still situations, however, where invoking the full LLM with hundreds of billions of parameters is too costly or inefficient. In those cases, EC uses the full LLM to generate training data for the task, then trains a much smaller, fine-tuned LLM from the foundation model and the training data to produce a highly-optimized and efficient model that meets use case accuracy requirements as well as cost and latency requirements.
Better Together
Google Cloud is a leader in the industry with massive natural language data sets and large-scale solutions. By partnering with Google Cloud, EC can leverage their expertise with LLMs and cloud-based solutions while focusing on the unique and differentiating AI technology they have developed to support deep content understanding and sophisticated logical reasoning.
Google Cloud has the reach to build large, accurate foundation models and host them reliably at scale and deliver them with speed.
Finally, Google has a strong research tradition that aligns very well with EC’s core values in research and innovation. This results in a unique partnership where EC and Google are willing to push the boundaries of the technology and work together to experiment, learn, and drive meaningful solutions for their customers.
Learn more about Google Cloud’s open and innovative generative AI partner ecosystem. Read more about the partnership between Elemental Cognition and Google Cloud here and request a demo here.
Cloud BlogRead More
