beginnerCertificate$29.99/mo

Deep Learning: Getting Started

Name: Deep Learning: Getting Started
Price: 29.99 USD
Rating: 4.4 (4500 reviews)

by Jonathan Fernandes · LinkedIn Learning

4.4

(4,500 reviews)

50K+ enrolled2 hoursUpdated 2024-04

Go to Course

Our Verdict

Worth it — with caveats

Deep Learning: Getting Started is worth taking only as a fast conceptual primer, not a comprehensive deep-learning course: in ~1 hour 13 minutes it gives newcomers (who already know some Python and ML) a clear mental model of neural networks plus three tiny Keras demos, and not much more. Taught by AI/ML leader Kumaran Ponnambalam (not Jonathan Fernandes, as listed in some catalogs), this LinkedIn Learning course holds a strong 4.6/5 rating from more than 2,000 learner ratings and walks through the conceptual building blocks of neural networks (perceptrons, weights and biases, activation functions, forward and back propagation, gradient descent) before implementing three short examples: Iris classification, spam text classification, and an incident root-cause-analysis exercise. LinkedIn officially labels it Intermediate, and the official 'Prerequisites for the course' lesson confirms it expects you to already know basic Python and the fundamentals of machine learning, so despite the 'Getting Started' title it is not built for true beginners.

It is a well-rated, clearly explained, low-cost (or free via the standard one-month LinkedIn Learning trial) conceptual primer that is genuinely good for absolute deep-learning newcomers who already have some Python and ML basics. But at roughly 73 minutes with only three small Keras demos, it is too shallow to make you job-ready or to satisfy anyone wanting mathematical depth or substantial coding practice, so it earns a conditional rather than an unqualified 'take.'

Best for: IT professionals, developers, analysts, and students who already know a little Python and basic machine learning and want a fast, jargon-light mental model of how neural networks work and how to build a minimal one in Keras before committing to a longer program. It also suits people who just need to 'speak the language' of deep learning for work conversations or to decide whether to invest in a deeper course.

Skip if: Complete programming beginners (it assumes Python and ML basics), people who want mathematical rigor or derivations, anyone seeking serious hands-on portfolio projects, and learners targeting modern topics like CNNs for vision, transformers/LLMs, or PyTorch. Those audiences will outgrow this 73-minute course almost immediately and should start with a more comprehensive specialization or university course.

About This Course

Build neural networks from scratch covering perceptrons, activation functions, backpropagation, and training with Keras.

What You'll Learn

What deep learning is and how it relates to linear regression and the perceptron

The architecture of an artificial neural network: input layer, hidden layers, weights and biases, activation functions, and output layer

How training works conceptually: forward propagation, measuring accuracy and error, back propagation, gradient descent, and batches and epochs

How to reuse existing network architectures and open-source pre-trained models (transfer-learning concept)

Building, training, evaluating, saving/loading, and predicting with a Keras deep-learning model on the Iris dataset

Creating text representations and building a spam-classification model on text data

Applying the workflow end-to-end in an incident root-cause-analysis exercise

Curriculum

Introduction

Getting started with deep learning, prerequisites for the course, and setting up the environment.

1. Introduction to Deep Learning

What is deep learning, linear regression, an analogy for deep learning, the perceptron, artificial neural networks, and training an ANN.

2. Neural Network Architecture

Input layer, hidden layers, weights and biases, activation functions, and the output layer.

3. Training a Neural Network

Setup and initialization, forward propagation, measuring accuracy and error, back propagation, gradient descent, batches and epochs, validation and testing, an ANN model, and reusing existing/open-source network architectures.

4. Deep Learning Example 1

Iris classification: input preprocessing, creating a deep-learning model in Keras, training and evaluation, saving and loading models, and making predictions.

5. Deep Learning Example 2

Spam classification: creating text representations, building a spam model, and making predictions for text.

6. Deep Learning Exercise

Incident root-cause analysis (RCA): problem statement, preprocessing RCA data, building the RCA model, and predicting root causes.

Conclusion

Extending your deep learning education.

Prerequisites

Basic Python programming familiarity
Fundamentals of machine learning (supervised learning, train/test concepts)
Ability to set up a Python/Jupyter environment with Keras/TensorFlow (the course includes a short 'Setting up the environment' lesson)

Instructor

Jonathan Fernandes

Instructor · LinkedIn Learning

Pros & Cons

Pros

Strong, independently verifiable rating of 4.6/5 from more than 2,000 ratings on LinkedIn Learning, with reviewers praising the simple, jargon-light explanations of neural-network building blocks
Very efficient: covers the full conceptual pipeline (perceptron through gradient descent) plus three runnable Keras examples in about 73 minutes
Genuinely hands-on for its length, with downloadable exercise files (Iris, spam, and root-cause-analysis notebooks) publicly mirrored on GitHub so you can follow along
Includes a shareable Certificate of Completion and is effectively free to finish within LinkedIn Learning's standard one-month free trial
Explanations are paced for non-mathematicians: each building block (perceptron, weights/biases, activation, back propagation, gradient descent) is introduced with a plain-language analogy before any code, which is why IT pros and analysts without a math background tend to finish it

Cons

Catalog metadata is inaccurate: the real instructor is Kumaran Ponnambalam (not Jonathan Fernandes) and the real runtime is about 1 hour 13 minutes (not 2 hours)
Far too short and shallow to make anyone job-ready; it is an orientation primer, not a complete deep-learning course, and skips the math
Content was published in November 2021 and omits modern essentials like CNNs for vision, transformers/LLMs, and PyTorch (it is Keras/TensorFlow only)
Despite the 'Getting Started' title, it assumes prior Python and machine-learning knowledge, so true beginners may struggle

Alternatives To Consider

Practical Deep Learning for Coders

fast.ai

View course

Deep Learning Specialization

Coursera

View course

PyTorch for Deep Learning & Machine Learning

Udemy

View course

Frequently Asked Questions

Is Deep Learning: Getting Started free?

Deep Learning: Getting Started is $29.99/mo. Requires a LinkedIn Learning subscription (about $29.99/month, or roughly $19.99/month billed annually). It can be completed for free within LinkedIn Learning's standard one-month free trial; many universities and public libraries also provide free access. No separate per-course purchase is offered.

Who is Deep Learning: Getting Started for?

IT professionals, developers, analysts, and students who already know a little Python and basic machine learning and want a fast, jargon-light mental model of how neural networks work and how to build a minimal one in Keras before committing to a longer program. It also suits people who just need to 'speak the language' of deep learning for work conversations or to decide whether to invest in a deeper course.

What will you learn in Deep Learning: Getting Started?

What deep learning is and how it relates to linear regression and the perceptron; The architecture of an artificial neural network: input layer, hidden layers, weights and biases, activation functions, and output layer; How training works conceptually: forward propagation, measuring accuracy and error, back propagation, gradient descent, and batches and epochs; How to reuse existing network architectures and open-source pre-trained models (transfer-learning concept).

What are the prerequisites for Deep Learning: Getting Started?

Basic Python programming familiarity; Fundamentals of machine learning (supervised learning, train/test concepts); Ability to set up a Python/Jupyter environment with Keras/TensorFlow (the course includes a short 'Setting up the environment' lesson).

Is Deep Learning: Getting Started worth it?

How we reviewed this course

This is an independent editorial assessment by Cursarium, based on LinkedIn Learning's published course materials and aggregated public learner feedback (last reviewed 2026-06). We have not independently completed the course. Links to providers are standard references, not paid placements.

Sources

$29.99/mo

Go to Course