Accenture has launched new services for testing artificial intelligence (AI) systems, powered by a ‘Teach and Test’ methodology designed to help companies build, monitor and measure reliable AI systems within their own infrastructure or in the cloud. The methodology ensures that AI systems are producing the right decisions in two phases. The ‘Teach’ phase focuses on the choice of data, models and algorithms that are used to train machine learning. This phase experiments and statistically evaluates different models to select the best performing model to be deployed into production, while avoiding gender, ethnic and other biases, as well as ethical and compliance risks.
Whereas, during the ‘Test’ phase, AI system outputs are compared to key performance indicators, and assessed for whether the system can explain how a decision or outcome was determined. It uses innovative techniques and cloud-based tools to monitor the system on an ongoing basis for sustained performance. For instance, a patent-pending normalization technique uses a unique algorithm to test object recognition more quickly.
“The adoption of AI is accelerating as businesses see its transformational value to power new innovations and growth. As organizations embrace AI, it is critical to find better ways to train and sustain these systems – securely and with quality – to avoid adverse effects on business performance, brand reputation, compliance and humans,” said Bhaskar Ghosh, Group Chief Executive, Accenture Technology Services.
According the Accenture Technology Vision 2018, AI systems require addressing many of the same challenges faced in human education and growth: fostering an understanding of right and wrong, and what it means to behave responsibly; imparting knowledge without bias; and building self-reliance while emphasizing the importance of collaborating and communicating with others.
“Testing AI systems presents a completely new set of challenges. While traditional application testing is deterministic, with a finite number of scenarios that can be defined in advance, AI systems require a limitless approach to testing. There is also a need for new capabilities for evaluating data and learning models, choosing algorithms, and monitoring for bias and ethical and regulatory compliance. Accenture’s ‘Teach and Test’ methodology takes all of this into consideration to help companies develop and validate AI systems with confidence,” said Kishore Durg, Senior Managing Director – Growth and Strategy and Global Testing Services Lead , Accenture.
Accenture used its ‘Teach and Test’ methodology to train a conversational virtual agent for a financial services company’s website, so that it could engage in accurate, unbiased conversations and know when to refer conversations to a human. The agent was trained 80 per cent faster than previously possible and achieved an 85 per cent accuracy rate on customer recommendations.
The methodology was also used to teach a sentiment analysis solution to evaluate a brand’s service performance by analyzing social media, news and other sources in real time. Having the right training data enabled the solution to correctly interpret sentiments in different contexts and domains. The training time for models was cut in half, enabling faster analysis and business results.
