AI Mentor for AI & Machine Learning Students

Every model you debug today is one less mystery in production tomorrow.

An engineering student asks Sumantrak™

“explain gradient descent and the learning rate for viva”

Gradient descent trains a model by repeatedly nudging its parameters in the direction that lowers the loss: it computes the gradient (the direction in which the loss rises fastest) and steps the opposite way. The learning rate is the step size: too large and the loss overshoots or even diverges, too small and training crawls or stalls on flat stretches. Batch, stochastic and mini-batch gradient descent differ only in how much data is used to estimate the gradient at each step. Viva tip: be ready to say why we subtract the gradient, and that on a non-convex loss you can land in a local minimum or a saddle point rather than the global minimum.

Free course

Sumantrak™ Fundamentals of AI

8 modules, 40 short lessons in English, Marathi or Hindi, an exam and a certificate. Open to AI & Machine Learning students in the first and second year.

Subjects Sumantrak™ covers for AI & Machine Learning

Business IntelligenceNatural Language EngineeringDigital Signal and Image ProcessingAI Knowledge Representation & ReasoningDeep LearningNatural Language ProcessingInformation Security and PrivacyMachine LearningJava ProgrammingData Engineering and Big Data SystemsData StructuresDesign and Analysis of AlgorithmsTheory of ComputationLinear Algebra

Also helps with GATE DA prep, same subject matter, exam-focused framing.

Set up my AI & Machine Learning mentor

Free to start, no card required.

Other branches