A Paris, France-and New York-based firm called Adaptive ML has secured $20 million in seed money to help businesses continuously enhance their generative AI models based on user interactions.
Index Ventures led the round, with participation from Factorial’s HuggingFund, Motier Ventures, Databricks Ventures, IRIS, and other angel investors.
With the first version of its enterprise platform now launched, the firm plans to use the funds to expand its workforce in Paris and New York and to keep investing in research and product development.
Adaptive, led by CEO Julien Launay, wants to enable businesses worldwide to use generative models that are always learning and provide intuitive user experiences in order to improve important business indicators.
The platform expedites the time-to-release for LLM-based products by abstracting away the technical details related to reinforcement learning and fine-tuning. Adaptive is designed for any kind of business that wants to use big language models, like software development, customer service, and in-depth research.
About Adaptive ML:
A more sophisticated approach that takes real-time data collection and analysis seriously is adaptive machine learning. It effortlessly adjusts to new information, as its name would imply, and offers insights practically instantly.
Adaptive machine learning (ML) uses a single channel as opposed to the two-channel or two-pipeline technique used in classical ML. Adaptive learning gathers and evaluates data in a gradual manner rather than all at once, in contrast to batch learning. This makes it possible for adaptive machine learning models to keep an eye on and absorb changes in input and output values. It also enables the model to modify its techniques for gathering, organizing, and analyzing data in response to fresh information.
Therefore, adaptive machine learning models will keep updating and modifying to give you the best predictors for future data as long as there is a stream of information coming in. You’ll get the best accuracy and maximum performance. The benefit of having a real-time system that is impervious to obsolescence is perhaps even more significant than the expense of maintaining AI technology.
