The relationship between the popular AI/deep learning models we use everyday and neuroscience research is not so obvious. Let’s explore some human cognitive skills that are serving as inspiration to a new generation of AI techniques.
Attention
Attention is one of those magical capabilities of the human brain that we don’t understand very well. What brain mechanisms allow us to focus on a specific task and ignore the rest of the environment? Attentional mechanisms have become a recent source of inspiration in deep learning models such as convolutional neural networks(CNNs) or deep generative models.
Episodic Memory
When you remember autobiographical events such as events or places we are using a brain function known as episodic memory. Recently, AI researchers have tried to incorporate methods inspired by episodic memory into reinforcement learning(RL) algorithms to episodic control. These networks store specific experiences (e.g., actions and reward outcomes associated with particular Atari game screens) and select new actions based on the similarity between the current situation input and the previous events stored in memory, taking the reward associated with those previous events into account.
Continual Learning
As humans we have the ability to learn new tasks without forgetting previous knowledge. Neural networks, in contrast suffer from what is known as the problem of catastrophic forgetting.
Inference
Human cognition is notorious for its ability to learn new concepts by drawing inspiration from previous knowledge through inductive inferences. Contrary to that, deep learning systems rely on massive amounts of training data to master the simplest of tasks. The rapidly growing field of meta-learning is another AI area of research inspired by the inference abilities of the human brain. The intersection of neuroscience and AI research is likely to power some of the most fascinating technology developments of the next decade. Somewhat counterintuitively, we should think about link between AI and neuroscience as a bi-directional relationship. While the human brain is certainly an inspiration for neural network architectures, advancements in AI are helping neuroscientists better understand obscure areas of the brain. Certainly, the new generation of neural networks is rapidly expanding beyond basic connections between neurons and recreating some of the core building blocks of human intelligence.
Imagination and Planning
Consciousness is related to the ability of humans( and other species) to forecast and think about the future. Most deep learning systems operate in incredibly reactive modes that makes it impossible to plan for longer term outcomes.
~ crossposted from KDNUGGETS
Comments