Image: Tensorleap
Development of neural networks is not a simple matter, which includes huge amounts of unstructured data and many hypotheses and experiments – expensive in time and costs. A new Israeli startup revealed today (Thursday) for the first time claims that it can repair and test your networks before the chaos begins.
Instead of trial and error – a real test
The Israeli startup Tensorleap is developing a platform capable of debugging and testing artificial neural networks (ANN) and deep neural networks (DNN) using algorithms. “The platform adds mathematical operations at each layer in the computational graph of the neural network in order to extract indicators from all the feature maps and evaluate their contribution to the decisions it makes,” explains Dodi Ben David, CEO of Tensorlip, in a conversation with Gigtime.
“Next, Tensorleap’s algorithms build the most informative space to explain the way the model interpreted the information, find clusters of data with similar characteristics, and more.” According to him, the accessibility of these analyzes to data scientists is a “Game Changer”: “Other tools mainly encourage the perpetuation of the existing paradigm of trial and error, and are not deep analysis tools for the models themselves. Therefore, their value is limited and they do not make a fundamental change in the development process or the quality of the developed model… A company like Tensorlip had to be established at some stage in order to allow the advancement of the field to the next stages, which is why we established it.’
With Gius Sid in pocket
Photography: Oren Dai
According to Ben David, a client from the field of genetic analysis estimated that a project he was developing would take about 9-10 months, but through the use of Tensorlip, it was completed in less than 3 months while finding bugs in the company’s information processing processes. The product is able to identify areas of weakness in the model, and understand why they fail. In addition, the system is able to identify balance problems in the databases used to train the model, and correct them. Before moving to production, the system performs dozens and hundreds of in-depth tests that make sure the model functions in any scenario it may encounter. “This is a fundamental difference from what exists today – a number of basic tests, which must pass a certain percentage in order to bring a model to production, which of course is not enough and leads to many failures later on,” says Ben David.
Tensorleap, founded in 2020, announced the completion of a $5.2 million seed round from Angular, Sozo and Industry, and currently employs approximately 15 people in its offices in Ramat Gan.