OE-11.2 Essentials of ANNs & Learning Strategies
You can compare neuron activation functions and ANN architectures, classify networks by connectivity and learning strategy, and apply the standard learning rules to numerical problems.
The activation function is what makes a network more than stacked linear algebra - without a non-linearity, any depth collapses to a single layer, and that is the sentence worth remembering. Supervised, unsupervised and reinforcement learning are distinguished by what feedback the network receives, not by network structure. Work the numerical problems by hand at least once; the weight-update arithmetic is what makes the learning rules concrete.
Work through these
Artificial neuron model and operations of the artificial neuron
The single unit in detail, and what it computes. Everything else in this subject is many of these arranged and trained.
Types of neuron activation function
The nonlinearity, and the reason it must be there at all. Without it a network of any depth collapses to a single linear map, which is the fact worth carrying.
ANN architectures
How units are arranged into layers and loops. The feedforward and recurrent split is the one that matters most.
Classification taxonomy of ANNs: connectivity
Sorting networks by how they are connected. It is a classification rather than a technique, and it makes reading the literature easier.
Learning strategies: supervised, unsupervised, reinforcement
The three ways a network can be told what to learn, distinguished by what feedback it gets. Choosing between them is decided by what data you actually have.
Learning rules
The rules that change the weights, which is where learning actually happens. Several are variations on the same idea of moving towards a better answer.
Numerical problems on weight updates
Working the arithmetic by hand for a few steps. It is tedious and it is the only reliable way to stop the process being magic.
Types of application
What this family gets used for, connecting back to the applications item in the previous topic.
Sign in to keep your progress.
Free resources
We haven't checked most of these for screen reader use yet.
Links last checked 30 Aug 2026.
Stuck here?
Ask a mentor. A real person answers, and they can see exactly which topic you're on. Usually within a couple of working days.
Checking your session…
Topics shown in module order.