Submission + - Biological neurons are for more complex than we imagined 1
Artem S. Tashkinov writes: Today, the most powerful artificial intelligence systems employ a type of machine learning called deep learning. Their algorithms learn by processing massive amounts of data through hidden layers of interconnected nodes, referred to as deep neural networks. As their name suggests, deep neural networks were inspired by the real neural networks in the brain, with the nodes modeled after real neurons — or, at least, after what neuroscientists knew about neurons back in the 1950s, when an influential neuron model called the perceptron was born. Since then, our understanding of the computational complexity of single neurons has dramatically expanded, so biological neurons are known to be more complex than artificial ones. But by how much?
To find out, David Beniaguev, Idan Segev and Michael London, all at the Hebrew University of Jerusalem, trained an artificial deep neural network to mimic the computations of a simulated biological neuron. They showed that a deep neural network requires between five and eight layers of interconnected “neurons” to represent the complexity of one single biological neuron. Even the authors did not anticipate such complexity. “I thought it would be simpler and smaller,” said Beniaguev. He expected that three or four layers would be enough to capture the computations performed within the cell.
To find out, David Beniaguev, Idan Segev and Michael London, all at the Hebrew University of Jerusalem, trained an artificial deep neural network to mimic the computations of a simulated biological neuron. They showed that a deep neural network requires between five and eight layers of interconnected “neurons” to represent the complexity of one single biological neuron. Even the authors did not anticipate such complexity. “I thought it would be simpler and smaller,” said Beniaguev. He expected that three or four layers would be enough to capture the computations performed within the cell.
Curious (Score:4, Interesting)
There are two types of computational neuron, the basic ones that are weighted inputs/weighted outputs that are not much more than N-way logic gates and the "biological simulator". If this is 8 deep on the biological simulators (such as Genesis 3) that's obviously far, far more significant than the more basic simulations. The article describes these as "simulated biological neurons", so I'm assuming the latter. They also say 8 layers of 256 neurons in each, so 2048 simulated neurons to one real neuron.
However, the preprint (https://www.biorxiv.org/content/10.1101/613141v2.full) suggests they used both the classic and bio sim methods. They also seem to have used home-grown code (available on github?) rather than a standard package, which has its good points and bad points as it means it's harder to find much written about how well the code works.