SEATTLE, Dec. 12, 2019 (GLOBE NEWSWIRE) — Algorithmia, a leader in machine learning deployment, announces the availability of its 2020 State of Enterprise Machine Learning report. The report includes input from 750 people across industries in companies that are actively engaged in building machine learning life cycles. A key takeaway from the report is that companies are increasing their machine learning investments, but challenges ranging from model deployment to scaling and testing persist.
In the last 12 months, there have been myriad developments in machine learning tools and applications. Computer vision is helping make queue lines more efficient in the UK, quantum computing for ML is gaining traction, and machine learning frameworks are releasing new iterations: PyTorch 1.0 and TensorFlow 2.0 are new in the last year. Meanwhile, hardware for AI and ML applications is also progressing. Google’s TPUs are in their third generation, the AWS Inferentia chip is one year old, Intel has Nervana Neural Network Processors, and Microsoft is reportedly developing its own custom AI hardware.
And yet, machine learning development remains in the early stages in most enterprises. Algorithmia’s research sought to identify and understand the reasons why.
Companies Are Looking to Machine Learning to Attract and Retain Customers
Undoubtedly, there are countless ways companies can apply machine learning to a particular business problem. For example, ML can be used to run prediction modeling to make assessments about customer churn, or to apply natural language processing to millions of tweets to analyze positive sentiment. In its 2020 survey, Algorithmia provides a wide-ranging list of possible ML use cases and a write-in options. Respondents were encouraged to select all answers that applied to their companies. The top five cases of companies of all sizes centered on increasing customer loyalty, led by generating customer insights and intelligence, and improving the customer experience, in addition to fraud detection, and reduction of overall costs.
Machine Learning Maturity is Increasing, But Deployment Challenges Remain
22% of respondents to Algorithmia’s 2020 survey said their companies have been in production with machine learning for a year. This might be considered good news, except most respondents (50 percent) said they spend between 8 and 90 days deploying a single machine learning model. Part of the problem appears to be scale, which 33% of survey respondents cited as the primary pain point in their machine learning life cycle.
Other pain points included version control and model reproducibility (32%), and getting executive buy-in (26%).
Companies Are Measuring ML Success with Business and Statistical Metrics
What does a successful ML program look like? Fifty-six percent of respondents said they are measuring ML success with business metrics, such as ROI, reduced customer churn, and product adoption. Simultaneously, the same number said they are using statistical metrics, including accuracy, precision, and speed, to measure success. Both metrics point to the importance of proof-of-concept in machine learning initiatives in order to prove the technology to senior management.
Now Is a Good Time to Have Machine Learning Experience On Your Resume
In 2016, Deloitte predicted a shortage of 180,000 data scientists by 2018. And from 2012 and 2017, the number of data scientist jobs on LinkedIn increased more than 650%. Clearly, ML expertise is in high demand—a point supported by Algorithmia’s 2020 study, which found that 44% of companies employ 10+ data scientists today, up from 17.5% in Algorithmia’s 2018 report. On the high end, Algorithmia’s 2020 report found that 5% of companies have more than 1,000 data scientists, whereas only 2% had as many in 2018.
“The findings of our 2020 study are consistent with what we’re hearing from customers,” said Diego Oppenheimer, CEO at Algorithmia. “Companies are growing their investments in machine learning, and machine learning operationalization is maturing across all industries, but significant room for growth and improvement remains. The model deployment lifecycle needs to continue to be more efficient and seamless for ML teams. Nevertheless, companies with established ML deployment lifecycles are benefiting from measurable results, including cost reductions, fraud detection, and customer satisfaction. We expect these trends to continue as ML technologies and processes arrive to market and are adopted.”
A copy of the complete Algorithmia 2020 State of Enterprise Machine Learning report can be found here.
Algorithmia’s software enables users to connect their data sources, orchestration engines, and step functions and deploy models from major frameworks, languages, platforms, and tools. We serve and scale model inference on infrastructure with high efficiency and empower users to manage their ML lifecycles with tools to collaborate, iterate, audit, secure, and govern. Over 90,000 engineers and data scientists have used Algorithmia’s platform to date, including the United Nations, numerous government intelligence agencies, and Fortune 500 companies. For more information, visit www.algorithmia.com.