In 1960, Nobel Laureate and American physicist Eugene Wigner wrote about the ‘unreasonable effectiveness of mathematics in natural sciences’. Mathematics is called the language of nature for a reason. That’s why the ‘Is math invented or discovered?’ debate never gets old. Mathematics exerts its influence on literally every field.

Mathematics is also the building block of machine learning models. ML practitioners use mathematics to analyse a problem, pick out better heuristics, and club both to generate an answer. Despite the critical role mathematics plays in machine learning, even state-of-art models struggle at maths.

A new study by the researchers at the University of California, Berkeley, have now introduced the MATH dataset. The team said the dataset provides a detailed assessment of a model’s mathematical ability across difficulties and subjects.

What Is MATH Dataset?

The MATH dataset consists of 12,500 problems taken from various high school mathematics competitions. The dataset measures the problem-solving ability of large and general-purpose language models. A machine learning model generates a sequence for a given problem from the MATH dataset and encodes the final answer. 

MATH problems are labelled from 1 to 5, depending on the difficulty level and span across seven subjects, including geometry, number theory, algebra calculus, statistics, and linear algebra. For problems of geometry, diagrams can be specified with the Asymptote language.

Since step-by-step solutions also accompany the problems, language models can learn to answer questions they haven’t been exposed to before. The step-by-step approach allows models to perform intermediate computations instead of giving the final answer immediately. 

Recognising the need to train the model on maths fundamentals before exposing to MATH that cover advanced problem-solving techniques, the team also released the Auxiliary Mathematics Problems and Solutions (AMPS). The ‘pretraining corpus’ has over 100,000 problems from Khan Academy with solutions and 5 million problems, based on 100 hand-designed modules, generated using Mathematica scripts. 

Results

When the MATH dataset was tested for large language models, including GPT-3, the accuracies were found to be abysmally low, ranging from 2.9 percent to 6.9 percent. However, on the flip side, the models achieved up to 15 percent accuracy on the easiest level. When evaluated on humans, a PhD student with no specialisation in Mathematics attained 40 percent, while a three-time Olympiad gold medalist scored 90 percent.

Further, having the models generate a step-by-step solution before producing the final answer reduced accuracy. This was because, while many of these steps were related to the question, they were not logical. 

The researchers found simply increasing the amount of training time, and the parameters proved extremely costly, although they did improve performance in a few cases. The researchers have open-sourced both MATH and AMPS to encourage and facilitate further research in this direction.

Predecessors

OpenAI recently introduced GPT-f, an automated prover and proof assistant for the Metamath formalisation language. Metamath is a language that expresses theorems in abstract mathematics along with proofs that a computer program can validate. 

Last year, Facebook built an AI system that can solve complex mathematical problems using symbolic reasoning. The team gave a system to represent mathematical expressions as a language and then treating the solutions as a translation problem for sequence-to-sequence neural networks.

Wrapping Up

While most other text-based tasks are already nearly solved by enormous Transformers, MATH is notably different. We showed that accuracy is slowly increasing and, if trends continue, the community will need to discover conceptual and algorithmic breakthroughs to attain strong performance on MATH, the researchers stated.

The post Domo Achieves AWS Machine Learning Competency Status in Applied AI appeared first on Artificial Intelligence.

Leading Mobile Cloud-based Business Operating system provider, Domo, is taking the Connected Devices route to bridge Sales and Marketing processes. In an exciting development for Marketing and Sales Technology aficionados, Domo has announced a strategic partnership with MarTech giant, Zendesk. This partnership with Zendesk will help Domo customers to better manage their IoT solutions and offer a better, proactive service experience to their end users.

How IoT App is a Value-Add to Domo Customers?

This new IoT app extends the Domo IoT Cloud provisions, enabling a better, proactive customer service experience with Zendesk. This is particularly focussed at creating a self-service autonomous workflow from device data to issue resolution.

Domo is already recognized as a leader in Self-Service BI Market by independent market research firms and analysts. This new IoT App further strengthens the Company’s BI product roadmap with Zendesk as a key partner.

Why you’ll love your Zendesk data in Domo

Modern customers expect an unprecedented level of experience delivered on every device they access. It’s easy to promise on Customer Experience dreams than to truly deliver on these. This Domo IoT App allows customers to move out of the vacuum created due to Marketing, Sales and Customer Service silos.

Intricately tying into every business operation, Zendesk’s Domo integration allow Domo customers to connect and manage all your other sources of customer data within Domo. This means you can get a holistic view of your customers’ needs and habits. With that kind of information, you won’t merely respond to helpdesk requests.

Role of Zendesk in Domo IoT App

With this new app, companies can merge their Zendesk data with data from other sources, all in the Domo platform. Once this data is in Domo, companies can get a holistic view of its customers’ needs and habits, and see how business decisions are affecting customers in real-time.

Benefits of Working with Zendesk within Domo

With the new IoT app, you can merge Zendesk data with data from any other source, all in one platform.

Additionally, you can create real-time alerts and receive instant notifications when Zendesk metrics hit thresholds that you determine.

The most satisfying benefit of having Zendesk work within Domo is the availability of Instant insights. Domo customers can now bring Zendesk and other data sources together to see how business decisions affect customers in real-time.

Marketing and Sales will Relish this Fluent Combination of CRM Data with IoT Functionality

Companies get real-time alerts based upon changes in the data coming from a device, and Domo pushes that data back to Zendesk Sunshine, the company’s CRM platform built on Amazon Web Services (AWS), to automatically initiate a ticket for Zendesk support. This app offers an end-to-end solution for companies to take action based upon the data from their IoT machines.

At the time of this announcement, Jay Heglar, Chief Strategy Officer at Domo., said, “Domo is helping customers evolve their IoT data from an operational tool for the shop floor to a valuable asset for the entire company.”

Jay added, “With this new App, we are providing our mutual customers with a seamless experience for taking data from machines and initiating (a) workflow to solve problems. Zendesk Sunshine makes it easy to deliver this integrated experience, which we believe is the future of IoT.”

Domo Customers Enjoy An Upper-Hand in Dealing with Heightened Customer Expectations

Tech does all the talking in this Zendesk+Domo IoT app function. Customers can leverage Domo’s new IoT App to merge an organization’s most relevant Zendesk data points and turns them into dynamic visualizations. These include:

Meet Customer Satisfaction Objectives

This feature allows users to easily analyze customer survey responses and breakdown the data by marketing channel, manager, location, teams, and individual agents, to identify trends and improve customer satisfaction scores.

Support Ticket Activity

This feature allows customers to view all of their support ticket data in one digestible dashboard, seeing opened and closed tickets, the number of escalations, and support team efficiency with a variety of filters and segments.

Autonomous Zendesk QuickStart

The Zendesk QuickStart provides a comprehensive summary of team performance, including backlog management, ticket handling, agent leaderboards, and much more. This feature allows customers to optimize their customer support and streamline their organization’s operations.

Norm Gennaro, SVP of worldwide sales at Zendesk, said, “The best customer experiences are built on Zendesk. Our customers are looking to drive excellent customer experiences and increased business value through IoT data, our new partnership with Domo helps us deliver on that opportunity.”

Norm added, “With the new Domo and Zendesk app, organizations can gain a holistic view of their customers’ needs and habits, and anticipate helpdesk requests that are affecting their customers in real-time.”

What Domo’s Loyal Customers Have to Say about  the new IoT Applications for Their Marketing, Sales, and Service

Companies like SharkNinja will be able to use this new solution to provide a new level of customer service to its install base.

“When SharkNinja launched the new Ion robot, they had hundreds of thousands of robots in consumers’ homes. And it would take them a whole week to analyze data from just 20 robots to learn things like ‘is the battery running low,’ or ‘is the robot completing its mission.’

Domo already offers its unique Data Science suite to customers. Adding its IoT App to the marketplace will heighten its sellability quotient, especially with Zendesk as its tech partner in the ecosystem.

Domo currently has over 30 IoT Integration Apps including AWS IoT Core, Azure IoT Hub, Raspberry Pi, and Soracom.

Besides the Zendesk integration, the newest in the fleet of IoT Integration Apps is the MQTT data and connectivity App. MQTT is a machine-to-machine, Internet of Things connectivity protocol. This protocol is lightweight and is suitable for constrained environments. This protocol is deployed in many IoT devices globally. MQTT is useful for connections with remote locations where a small code footprint is required and/or network bandwidth is at a premium.

For example, MQTT has been used in sensors communicating to a broker via satellite link, over occasional dial-up connections with healthcare providers, and in a range of home automation and small device scenarios. It is also ideal for mobile applications because of its small size, low power usage, minimized data packets, and efficient distribution of information to one or many receivers.

Currently, Domo’s mission is to be the operating system for business, digitally connecting all your people, your data and your systems, empowering them to collaborate better, make better decisions and be more efficient, right from their phones. Domo works with many of the world’s leading and most progressive brands across multiple industries including retail, media and entertainment, manufacturing, finance and more.

The post Domo IoT Cloud Now Integrates with Zendesk Data Platform to Deliver Better Customer Service appeared first on Artificial Intelligence.