Fine-Tuning

Fine-tuning is the process of adjusting an artificial intelligence (AI) solution’s parameters to optimize outputs or help the solution adapt to a new context.

What Is Fine-Tuning?

Training an AI involves starting from the ground up by feeding the solution data sets to help it develop its core processing capabilities. Fine-tuning, on the other hand, involves tweaking an existing AI model’s parameters to meet a changing goal. Say a brand shifts its core messaging and product descriptions. The brand would then need to ensure its AI chatbot and AI-powered search engine reflect the latest products and brand messaging. In this case, developers would adjust the model’s parameters, feed it new data, and monitor the AI’s outputs to ensure it’s performing accurately to the new brand goals. Fine-tuning can also be useful for ongoing progress checks. Because AI solutions rely on available data and information to form results, their results can shift as that information evolves. Fine-tuning helps ensure answers and outputs are always accurate within a shifting data environment.

See Salsify in Action

Bring your team, tools, and data together to launch high-impact inbound marketing campaigns in less time.

Fine-Tuning

What Is Fine-Tuning?

Related Terms

Related Resources

PXM AI Manifesto: How AI Transforms Product Experience Management

Under the Hood: Salsify AI Features That Will Revolutionize PXM

See Salsify in Action