Extreme learning machine: a new learning scheme of feedforward neural networks

TLDR

Feedforward neural networks suffer from slow learning speeds, largely due to reliance on gradient‑based algorithms that iteratively tune all parameters, creating a major bottleneck. The study introduces the extreme learning machine (ELM) to accelerate training of single‑hidden layer feedforward neural networks by randomly assigning input weights and analytically computing output weights. ELM trains SLFNs by randomly selecting input weights and analytically solving for output weights, bypassing iterative gradient‑based updates. The ELM achieves superior generalization and markedly faster learning than traditional algorithms, as demonstrated on real‑world function approximation and classification benchmarks.

Abstract

It is clear that the learning speed of feedforward neural networks is in general far slower than required and it has been a major bottleneck in their applications for past decades. Two key reasons behind may be: 1) the slow gradient-based learning algorithms are extensively used to train neural networks, and 2) all the parameters of the networks are tuned iteratively by using such learning algorithms. Unlike these traditional implementations, this paper proposes a new learning algorithm called extreme learning machine (ELM) for single-hidden layer feedforward neural networks (SLFNs) which randomly chooses the input weights and analytically determines the output weights of SLFNs. In theory, this algorithm tends to provide the best generalization performance at extremely fast learning speed. The experimental results based on real-world benchmarking function approximation and classification problems including large complex applications show that the new algorithm can produce best generalization performance in some cases and can learn much faster than traditional popular learning algorithms for feedforward neural networks.