Abrupt and spontaneous strategy switches emerge in simple regularised neural networks

• URL: http://arxiv.org/abs/2302.11351v4
• Date: Fri, 1 Mar 2024 16:54:07 GMT
• Title: Abrupt and spontaneous strategy switches emerge in simple regularised neural networks
• Authors: Anika T. L\"owe, L\'eo Touzo, Paul S. Muhle-Karbe, Andrew M. Saxe, Christopher Summerfield, Nicolas W. Schuck
• Abstract summary: We study whether insight-like behaviour can occur in simple artificial neural networks. Analyses of network architectures and learning dynamics revealed that insight-like behaviour crucially depended on a regularised gating mechanism. This suggests that insight-like behaviour can arise naturally from gradual learning in simple neural networks.
• Score: 8.737068885923348
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Humans sometimes have an insight that leads to a sudden and drastic performance improvement on the task they are working on. Sudden strategy adaptations are often linked to insights, considered to be a unique aspect of human cognition tied to complex processes such as creativity or meta-cognitive reasoning. Here, we take a learning perspective and ask whether insight-like behaviour can occur in simple artificial neural networks, even when the models only learn to form input-output associations through gradual gradient descent. We compared learning dynamics in humans and regularised neural networks in a perceptual decision task that included a hidden regularity to solve the task more efficiently. Our results show that only some humans discover this regularity, whose behaviour was marked by a sudden and abrupt strategy switch that reflects an aha-moment. Notably, we find that simple neural networks with a gradual learning rule and a constant learning rate closely mimicked behavioural characteristics of human insight-like switches, exhibiting delay of insight, suddenness and selective occurrence in only some networks. Analyses of network architectures and learning dynamics revealed that insight-like behaviour crucially depended on a regularised gating mechanism and noise added to gradient updates, which allowed the networks to accumulate "silent knowledge" that is initially suppressed by regularised (attentional) gating. This suggests that insight-like behaviour can arise naturally from gradual learning in simple neural networks, where it reflects the combined influences of noise, gating and regularisation.

Related papers

• Life, uh, Finds a Way: Systematic Neural Search [2.163881720692685]
  We tackle the challenge of rapidly adapting an agent's behavior to solve continuous problems in settings. Instead of focusing on deep reinforcement learning, we propose viewing behavior as the physical manifestation of a search procedure. We describe an algorithm that implicitly enumerates behaviors by regulating the tight feedback loop between execution of behaviors and mutation of the graph.
  arXiv  Detail & Related papers  (2024-10-02T09:06:54Z)
• Contrastive-Signal-Dependent Plasticity: Self-Supervised Learning in Spiking Neural Circuits [61.94533459151743]
  This work addresses the challenge of designing neurobiologically-motivated schemes for adjusting the synapses of spiking networks. Our experimental simulations demonstrate a consistent advantage over other biologically-plausible approaches when training recurrent spiking networks.
  arXiv  Detail & Related papers  (2023-03-30T02:40:28Z)
• Critical Learning Periods for Multisensory Integration in Deep Networks [112.40005682521638]
  We show that the ability of a neural network to integrate information from diverse sources hinges critically on being exposed to properly correlated signals during the early phases of training. We show that critical periods arise from the complex and unstable early transient dynamics, which are decisive of final performance of the trained system and their learned representations.
  arXiv  Detail & Related papers  (2022-10-06T23:50:38Z)
• Searching for the Essence of Adversarial Perturbations [73.96215665913797]
  We show that adversarial perturbations contain human-recognizable information, which is the key conspirator responsible for a neural network's erroneous prediction. This concept of human-recognizable information allows us to explain key features related to adversarial perturbations.
  arXiv  Detail & Related papers  (2022-05-30T18:04:57Z)
• The world seems different in a social context: a neural network analysis of human experimental data [57.729312306803955]
  We show that it is possible to replicate human behavioral data in both individual and social task settings by modifying the precision of prior and sensory signals. An analysis of the neural activation traces of the trained networks provides evidence that information is coded in fundamentally different ways in the network in the individual and in the social conditions.
  arXiv  Detail & Related papers  (2022-03-03T17:19:12Z)
• Data-driven emergence of convolutional structure in neural networks [83.4920717252233]
  We show how fully-connected neural networks solving a discrimination task can learn a convolutional structure directly from their inputs. By carefully designing data models, we show that the emergence of this pattern is triggered by the non-Gaussian, higher-order local structure of the inputs.
  arXiv  Detail & Related papers  (2022-02-01T17:11:13Z)
• Learning to acquire novel cognitive tasks with evolution, plasticity and meta-meta-learning [3.8073142980733]
  In meta-learning, networks are trained with external algorithms to learn tasks that require acquiring, storing and exploiting unpredictable information for each new instance of the task. Here we evolve neural networks, endowed with plastic connections, over a sizable set of simple meta-learning tasks based on a neuroscience modelling framework. The resulting evolved network can automatically acquire a novel simple cognitive task, never seen during training, through the spontaneous operation of its evolved neural organization and plasticity structure.
  arXiv  Detail & Related papers  (2021-12-16T03:18:01Z)
• Gradient Starvation: A Learning Proclivity in Neural Networks [97.02382916372594]
  Gradient Starvation arises when cross-entropy loss is minimized by capturing only a subset of features relevant for the task. This work provides a theoretical explanation for the emergence of such feature imbalance in neural networks.
  arXiv  Detail & Related papers  (2020-11-18T18:52:08Z)
• Relationship between manifold smoothness and adversarial vulnerability in deep learning with local errors [2.7834038784275403]
  We study the origin of the adversarial vulnerability in artificial neural networks. Our study reveals that a high generalization accuracy requires a relatively fast power-law decay of the eigen-spectrum of hidden representations.
  arXiv  Detail & Related papers  (2020-07-04T08:47:51Z)
• Adaptive Reinforcement Learning through Evolving Self-Modifying Neural Networks [0.0]
  Current methods in Reinforcement Learning (RL) only adjust to new interactions after reflection over a specified time interval. Recent work addressing this by endowing artificial neural networks with neuromodulated plasticity have been shown to improve performance on simple RL tasks trained using backpropagation. Here we study the problem of meta-learning in a challenging quadruped domain, where each leg of the quadruped has a chance of becoming unusable. Results demonstrate that agents evolved using self-modifying plastic networks are more capable of adapting to complex meta-learning learning tasks, even outperforming the same network updated using gradient
  arXiv  Detail & Related papers  (2020-05-22T02:24:44Z)
• How Do You Act? An Empirical Study to Understand Behavior of Deep Reinforcement Learning Agents [2.3268634502937937]
  The demand for more transparency of decision-making processes of deep reinforcement learning agents is greater than ever. In this study, we characterize the learned representations of an agent's policy network through its activation space. We show that the healthy agent's behavior is characterized by a distinct correlation pattern between the network's layer activation and the performed actions.
  arXiv  Detail & Related papers  (2020-04-07T10:08:55Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.