Cambridge Quantum pushes into NLP and quantum computing with new head of AI

Explore the future of AI on August 5 in San Francisco—join Block, GSK, and SAP at Autonomous Workforces to discover how enterprises are scaling multi-agent systems with real-world results. Join leaders from Block, GSK, and SAP for an exclusive look at how autonomous agents are reshaping enterprise workflows – from real-time decision-making to end-to-end automation. AI will almost certainly see rapid growth over the coming years as it continues to disrupt customer service.

Cost

Study characteristics included were the number of participants, age, gender, and NLP information, such as task assigned, ground truth (gold standard), and the type of NLP algorithms used. Transformer models take applications such as language translation and chatbots to a new level. Innovations such as the self-attention mechanism and multi-head attention enable these models to better weigh the importance of various parts of the input, and to process those parts in parallel rather than sequentially. Despite these limitations, the researchers believe the study was able to reach moderate-to-high performance levels in selecting patients who may benefit from epilepsy surgery, when other treatments are no longer an appropriate option for them. Rajeswaran V, senior director at Capgemini, notes that Open AI’s GPT-3 model has mastered language without using any labeled data.

• By relying on morphology — the study of words, how they are formed, and their relationship to other words in the same language — GPT-3 can perform language translation much better than existing state-of-the-art models, he says.
• One criterion for the test involved deciding whether the computer could interpret and generate natural language.
• For example, amid concerns that improvements in quantum hardware will make it easier to break existing algorithms used in modern cryptography, CQC devised a method to generate quantum-resistant cryptographic keys that cannot be cracked by today’s methods.

Studies suggest natural language processing (NLP) is an effective tool in identifying candidates for epilepsy surgery. Cambridge Quantum Computing (CQC) hiring Stephen Clark as head of AI last week could be a sign the company is boosting research into ways quantum computing could be used for natural language processing. For example, tools like Generative Pre-trained Transformer 3 (GPT-3), developed by OpenAI, use a neural network machine learning model that can not only code but also write articles and answer questions, frequently in a manner virtually indistinguishable from a human response.

NLP is about getting computers to perform useful and interesting tasks involving spoken and written human language. NLP is sometimes referred to as Computational Linguistics to emphasize the fact that involves the combination of CS methods with research insights from Linguistics (the study of human language). Practical applications of NLP include question answering, machine translation, information extraction, and interactive dialog systems (both written and spoken). Modern NLP systems rely heavily on methods involving probability, linear algebra and calculus often in combination with machine learning methods. NLP leverages machine learning (ML) algorithms trained on unstructured data, typically text, to analyze how elements of human language are structured together to impart meaning. Phrases, sentences, and sometimes entire books are fed into ML engines where they’re processed using grammatical rules, people’s real-life linguistic habits, and the like.

Advertise with MIT Technology Review

• Because they require a lot of context-specific training, they’re also not flexible.
• Although Xena may never be able to clear out the refrigerator in your office building or ensure everyone actually signs a birthday card, the agent is likely a harbinger of bigger things to come in the NLP world.
• Many of the features of an NLP engine, such as named entities, sentence grammar and speaker intention, are all provided by machine learning models, which are foundational to what we call “AI” today.
• You’ve probably benefitted from NLP if you’ve asked your smartphone for directions to a restaurant or run a spell-check on a word processing program.
• Clark was previously senior staff research scientist at DeepMind and led a team working on grounded language learning in virtual environments.

If you want to impress your boss, VB Daily has you covered. We give you the inside scoop on what companies are doing with generative AI, from regulatory shifts to practical deployments, so you can share insights for maximum ROI. Although Xena may never be able to clear out the refrigerator in your office building or ensure everyone actually signs a birthday card, the agent is likely a harbinger of bigger things to come in the NLP world. Smart companies are already considering how to utilize NLP and other AI tools to make their workplaces more efficient and profitable. And smart investors will pay attention to these tools and how they’re used as they continue to develop. The start-up Xembly is using an automated, NLP-powered platform to handle many office jobs that often get lost in the shuffle.

Development of natural language processing

By relying on morphology — the study of words, how they are formed, and their relationship to other words in the same language — GPT-3 can perform language translation much better than existing state-of-the-art models, he says. These algorithms give machines the ability to answer complex and nuanced questions. While the impressive results are a remarkable leap beyond what existing language models have achieved, the technique involved isn’t exactly new. Instead, the breakthrough was driven primarily by feeding the algorithm ever more training data—a trick that has also been responsible for most of the other recent advancements in teaching AI to read and write. “It’s kind of surprising people in terms of what you can do with … more data and bigger models,” says Percy Liang, a computer science professor at Stanford. Examples of these problems include intentions (i.e., whether a customer intends to buy, sell, cancel or recommend, based on their feedback) and effort (how hard it is for a customer to complete an action with an organization’s product or service).

NLP isn’t going anywhere and will likely become one of the cornerstones of a company’s AI philosophy and plan. They, too, are adding to NLP/NLU systems with features like automated content summaries that help answer business questions like, “Why did the sentiment around my new product dip in Q2? ” LLMs like ChatGPT or local LLMs like Llama are ideal for digging into a collection of documents and answering the “why” and “what to do about it” questions.

Other ways to search:

In the short term, Liang thinks, the field of NLP will see much more progress from exploiting existing techniques, particularly those based on distributional semantics. “There’s probably a qualitative gap between the way that humans understand language and perceive the world and our current models,” he says. Closing that gap would probably require a new way of thinking, he adds, as well as much more time.

Natural language processing innovation

An NLP algorithm uses this data to find patterns and extrapolate what comes next. For example, a translation algorithm that recognizes that, in French, “I’m going to the park” is “Je vais au parc” will learn to predict that “I’m going to the store” also begins with “Je vais au.” All the algorithm then needs is the word for “store” to complete the translation task. The addition of Clark to CQC’s team signals the company will be shifting some of its research and development efforts toward quantum natural language processing (QNLP). Humans are good at composing meanings, but this process is not well understood.

Its conversational AI agent, Xena, can listen to meetings, take notes, schedule meetings through Slack or email, remind people of action items, and even understand who is talking to whom when there is more than one speaker. The result was a shift to statistical methods to develop NLP capabilities. The current approach to NLP uses both linguistic and statistical methods to interpret and respond to instructions. For the next 50 years, linguists developed NLP using painstaking trial-and-error rules. In the 1990s, however, computers became much faster and more capable of doing calculations in seconds, even those that previously took hours or days.