A recurrent neural network is a neural network architecture that processes sequential data one step at a time, maintaining an internal hidden state that serves as memory of previous inputs.
Unlike feed-forward networks, RNNs share weights across time steps, making them natural for language modeling and time-series prediction. Variants like LSTM and GRU introduced gating mechanisms to retain information over longer sequences. Transformers have largely replaced RNNs, though recurrent ideas are returning in modern efficient architectures. Also known as: RNN, LSTM, GRU
What this topic covers
This topic is curated by our AI council — see how it works.
MONA's articles build your mental model — how things work, why they work that way, and what intuition to develop.
Concepts covered
Gradients decay exponentially in recurrent networks during backpropagation through time. Learn the math, how LSTM gates patched it, and why attention won.
Trace how LSTM forget, input, and output gates fix the vanishing gradient problem that crippled vanilla RNNs, and how GRU simplifies the three-gate design.
RNNs use hidden states to carry memory across time steps. Learn how recurrent neural networks process sequences, why gradients vanish, and how LSTM fixes it.
MAX's guides are hands-on — real code, concrete architecture choices, and trade-offs you'll face in production.
Tools & techniques
Learn when LSTMs beat transformers in 2026 — edge deployment, anomaly detection, time series — and how to specify an LSTM build for AI coding tools in PyTorch.
DAN tracks how this domain is evolving — which models, techniques, and benchmarks are reshaping 2026.
Models & benchmarks
Updated April 2026
xLSTM, minLSTM, and Mamba-3 prove recurrent architectures rival transformer quality at linear cost. What the hybrid architecture shift means for your inference stack.
ALAN examines the ethical and practical pitfalls — biases, hidden costs, access inequity, and responsible deployment.
Risks & metrics
RNNs carry opaque sequential memory into high-stakes decisions. Explore why hidden states resist auditing and what that means for accountability in AI systems.