Categorical Perception: A Groundwork for Deep Learning

Bonnasse‐Gahot, Laurent; Nadal, Jean‐Pierre

doi:10.1162/neco_a_01454

Cited by 5 publications

(2 citation statements)

References 61 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some theoretical models have been proposed to explain how the category is learned in the neural system ( 26 – 28 ). But most of models show categorical phenomena that are consistent with some behavioral results, without showing neural activity that encodes category information observed in the PFC or other brain areas ( 29 , 30 ). Hinaut and Dominey constructed a neural network model of the PFC that demonstrated how categorization of behavioral sequences can be achieved through a recurrent system ( 31 ).…”

Section: Introductionmentioning

confidence: 61%

Category learning in a recurrent neural network with reinforcement learning

Zhang

Pan²,

Wang³

2022

Front. Psychiatry

View full text Add to dashboard Cite

It is known that humans and animals can learn and utilize category information quickly and efficiently to adapt to changing environments, and several brain areas are involved in learning and encoding category information. However, it is unclear that how the brain system learns and forms categorical representations from the view of neural circuits. In order to investigate this issue from the network level, we combine a recurrent neural network with reinforcement learning to construct a deep reinforcement learning model to demonstrate how the category is learned and represented in the network. The model consists of a policy network and a value network. The policy network is responsible for updating the policy to choose actions, while the value network is responsible for evaluating the action to predict rewards. The agent learns dynamically through the information interaction between the policy network and the value network. This model was trained to learn six stimulus-stimulus associative chains in a sequential paired-association task that was learned by the monkey. The simulated results demonstrated that our model was able to learn the stimulus-stimulus associative chains, and successfully reproduced the similar behavior of the monkey performing the same task. Two types of neurons were found in this model: one type primarily encoded identity information about individual stimuli; the other type mainly encoded category information of associated stimuli in one chain. The two types of activity-patterns were also observed in the primate prefrontal cortex after the monkey learned the same task. Furthermore, the ability of these two types of neurons to encode stimulus or category information was enhanced during this model was learning the task. Our results suggest that the neurons in the recurrent neural network have the ability to form categorical representations through deep reinforcement learning during learning stimulus-stimulus associations. It might provide a new approach for understanding neuronal mechanisms underlying how the prefrontal cortex learns and encodes category information.

show abstract

Section: Introductionmentioning

confidence: 61%

Category learning in a recurrent neural network with reinforcement learning

Zhang

Pan²,

Wang³

2022

Front. Psychiatry

View full text Add to dashboard Cite

show abstract

“…The cited studies (Thériault et al, 2018;Bonnasse-Gahot & Nadal, 2022) suggest that deep learning models can exhibit a form of CP at the level of their internal representations.…”

Section: Categorical Perception In Neural Networkmentioning

confidence: 99%

Language Writ Large: LLMs, ChatGPT, Grounding, Meaning and Understanding

Harnad

2024

Preprint

View full text Add to dashboard Cite

Apart from what (little) OpenAI may be concealing from us, we all know (roughly) how ChatGPT works (its huge text database, its statistics, its vector representations, and their huge number of parameters, its next-word training, etc.). But none of us can say (hand on heart) that we are not surprised by what ChatGPT has proved to be able to do with these resources. This has even driven some of us to conclude that ChatGPT actually understands. It’s not true that it understands. But it is also not true that we understand how it can do what it can do. I will suggest some hunches about benign “biases”: convergent constraints that emerge at LLM-scale that may be helping ChatGPT do so much better than we would have expected. These biases are inherent in the nature of language itself, at LLM-scale, and they are closely linked to what it is that ChatGPT lacks, which is direct sensorimotor grounding to connect its words to their referents and its propositions to their meanings. These convergent biases are related to (1) the parasitism of indirect verbal grounding on direct sensorimotor grounding, (2) the circularity of verbal definition, (3) the “mirroring” of language production and comprehension, (4) iconicity in propositions at LLM-scale, (5) computational counterparts of human “categorical perception” in category learning by neural nets, and perhaps also (6) a conjecture by Chomsky about the laws of thought. The exposition will be in the form of a dialogue with ChatGPT-4.

show abstract

Evolution of neural activity in circuits bridging sensory and abstract knowledge

Mastrogiuseppe

Hiratani

Latham

2022

Preprint

View full text Add to dashboard Cite

The ability to associate sensory stimuli with abstract classes is critical for survival. How are these associations implemented in brain circuits? And what governs how neural activity evolves during abstract knowledge acquisition? To investigate these questions, we consider a circuit model that learns to map sensory inputs into abstract classes via gradient descent synaptic plasticity. We focus on typical neuroscience tasks (simple, and context-dependent, categorization), and study how both synaptic connectivity and neural activity evolve during learning. To make contact with the current generation of experiments we focus on the latter, and analyze activity via standard measures such as selectivity, correlations, and tuning symmetry. We find that the model is able to capture experimental observations, including seemingly disparate ones. We determine how, in the model, the behaviour of these activity measures depends on details of the circuit and the task. These dependencies make experimentally-testable predictions about the circuitry supporting abstract knowledge acquisition in the brain.

show abstract

Categorical Perception: A Groundwork for Deep Learning

Cited by 5 publications

References 61 publications

Category learning in a recurrent neural network with reinforcement learning

Category learning in a recurrent neural network with reinforcement learning

Language Writ Large: LLMs, ChatGPT, Grounding, Meaning and Understanding

Evolution of neural activity in circuits bridging sensory and abstract knowledge

Contact Info

Product

Resources

About