Search for: All records

Abstract Distinguishing between nectar and non-nectar odors is challenging for animals due to shared compounds and varying ratios in complex mixtures. Changes in nectar production throughout the day and over the animal’s lifetime add to the complexity. The honeybee olfactory system, containing fewer than 1000 principal neurons in the early olfactory relay, the antennal lobe (AL), must learn to associate diverse volatile blends with rewards. Previous studies identified plasticity in the AL circuits, but its role in odor learning remains poorly understood. Using a biophysical computational model, tuned by in vivo electrophysiological data, and live imaging of the honeybee’s AL, we explored the neural mechanisms of plasticity in the AL. Our findings revealed that when trained with a set of rewarded and unrewarded odors, the AL inhibitory network suppresses responses to shared chemical compounds while enhancing responses to distinct compounds. This results in improved pattern separation and a more concise neural code. Our calcium imaging data support these predictions. Analysis of a graph convolutional neural network performing an odor categorization task revealed a similar mechanism for contrast enhancement. Our study provides insights into how inhibitory plasticity in the early olfactory network reshapes the coding for efficient learning of complex odors. 

Migratory locusts (Locusta migratoria) emit two key odorants during aggregation: 4-vinylanisole (4VA), which serves as an aggregation pheromone attracting conspecifics to form swarms, and phenylacetonitrile (PAN), which acts as an aposematic signal and a precursor of a defense toxin, deterring conspecifics from cannibalism and protecting against predators. However, how locusts reconcile these two conflicting olfactory signals while aggregating is not yet understood. Our study addresses this by examining the release dynamics of the two signals, their behavioral effects, and the neural mechanisms underlying their perception. 4VA is released earlier and at lower locust densities than PAN, with PAN’s release increasing as aggregation progresses. Although PAN’s emission levels eventually exceed those of 4VA, locusts consistently exhibit a preference for the emitted blend, regardless of variations in proportions and concentrations. Notably, increasing amounts of 4VA added to PAN can counteract PAN’s repellent effects, but this is not the case when PAN is added to 4VA. Mechanistically, we found that antennal neurons responsive to 4VA suppress the activity of neurons responsive to PAN. In the antennal lobe, it is the conduction velocities of projection neurons, rather than other neural properties, that are responsible for the observed behavioral pattern, leading to an overall attractive response. Collectively, our findings imply that insects are capable of harmonizing the effects of two distinct pheromones to optimize both social cohesion and chemical defense. 

Artificial neural networks (ANNs) struggle with continual learning, sacrificing performance on previously learned tasks to acquire new task knowledge. Here we propose a new approach allowing to mitigate catastrophic forgetting during continuous task learning. Typically a new task is trained until it reaches maximal performance, causing complete catastrophic forgetting of the previous tasks. In our new approach, termed Optimal Stopping (OS), network training on each new task continues only while the mean validation accuracy across all the tasks (current and previous) increases. The stopping criterion creates an explicit balance: lower performance on new tasks is accepted in exchange for preserving knowledge of previous tasks, resulting in higher overall network performance. The overall performance is further improved when OS is combined with Sleep Replay Consolidation (SRC), wherein the network converts to a Spiking Neural Network (SNN) and undergoes unsupervised learning modulated by Hebbian plasticity. During the SRC, the network spontaneously replays activation patterns from previous tasks, helping to maintain and restore prior task performance. This combined approach offers a promising avenue for enhancing the robustness and longevity of learned representations in continual learning models, achieving over twice the mean accuracy of baseline continuous learning while maintaining stable performance across tasks. 

When trained on biased datasets, Deep Neural Networks (DNNs) often make predictions based on attributes derived from features spuriously correlated with the target labels. This is especially problematic if these irrelevant features are easier for the model to learn than the truly relevant ones. Many existing approaches, called debiasing methods, have been proposed to address this issue, but they often require predefined bias labels and entail significantly increased computational complexity by incorporating extra auxiliary models. Instead, we provide an orthogonal perspective from the existing approaches, inspired by cognitive science, specifically Global Workspace Theory (GWT). Our method, Debiasing Global Workspace (DGW), is a novel debiasing framework that consists of specialized modules and a shared workspace, allowing for increased modularity and improved debiasing performance. Additionally, DGW enhances the transparency of decision-making processes by visualizing which features of the inputs the model focuses on during training and inference through attention masks. We begin by proposing an instantiation of GWT for the debiasing method. We then outline the implementation of each component within DGW. At the end, we validate our method across various biased datasets, proving its effectiveness in mitigating biases and improving model performance. 

Artificial neural networks (ANNs) show limited performance with scarce or imbalanced training data and face challenges with continuous learning, such as forgetting previously learned data after new tasks training. In contrast, the human brain can learn continuously and from just a few examples. This research explores the impact of ’sleep’ an unsupervised phase incorporating stochastic network activation with local Hebbian learning rules on ANNs trained incrementally with limited and imbalanced datasets, specifically MNIST and Fashion MNIST. We discovered that introducing a sleep phase significantly enhanced accuracy in models trained with limited data. When a few tasks were trained sequentially, sleep replay not only rescued previously learned information that had been forgotten following new task training but also often enhanced performance in prior tasks, especially those trained with limited data. This study highlights the multifaceted role of sleep replay in augmenting learning efficiency and facilitating continual learning in ANNs. 

The performance of artificial neural networks (ANNs) degrades when training data are limited or imbalanced. In contrast, the human brain can learn quickly from just a few examples. Here, we investigated the role of sleep in improving the performance of ANNs trained with limited data on the MNIST and Fashion MNIST datasets. Sleep was implemented as an unsupervised phase with local Hebbian type learning rules. We found a significant boost in accuracy after the sleep phase for models trained with limited data in the range of 0.5-10% of total MNIST or Fashion MNIST datasets. When more than 10% of the total data was used, sleep alone had a slight negative impact on performance, but this was remedied by fine-tuning on the original data. This study sheds light on a potential synaptic weight dynamics strategy employed by the brain during sleep to enhance memory performance when training data are limited or imbalanced. 

Many interpretable AI approaches have been proposed to provide plausible explanations for a model’s decision-making. However, configuring an explainable model that effectively communicates among computational modules has received less attention. A recently proposed shared global workspace theory showed that networks of distributed modules can benefit from sharing information with a bottle-necked memory because the communication constraints encourage specialization, compositionality, and synchronization among the modules. Inspired by this, we propose Concept-Centric Transformers, a simple yet effective configuration of the shared global workspace for interpretability, consisting of: i) an object-centric-based memory module for extracting semantic concepts from input features, ii) a cross-attention mechanism between the learned concept and input embeddings, and iii) standard classification and explanation losses to allow human analysts to directly assess an explanation for the model’s classification reasoning. We test our approach against other existing concept-based methods on classification tasks for various datasets, including CIFAR100, CUB-200-2011, and ImageNet, and we show that our model achieves better classification accuracy than all baselines across all problems but also generates more consistent concept-based explanations of classification output. 

Convolutional neural networks (CNNs) are a foundational model architecture utilized to perform a wide variety of visual tasks. On image classification tasks CNNs achieve high performance, however model accuracy degrades quickly when inputs are perturbed by distortions such as additive noise or blurring. This drop in performance partly arises from incorrect detection of local features by convolutional layers. In this work, we develop a neuroscience-inspired unsupervised Sleep Replay Consolidation (SRC) algorithm for improving convolutional filter’s robustness to perturbations. We demonstrate that sleep- based optimization improves the quality of convolutional layers by the selective modification of spatial gradients across filters. We further show that, compared to other approaches such as fine- tuning, a single sleep phase improves robustness across different types of distortions in a data efficient manner. 

Geometric Sensitive Hashing functions, a family of Local Sensitive Hashing functions, are neural network models that learn class-specific manifold geometry in supervised learning. However, given a set of supervised learning tasks, understanding the manifold geometries that can represent each task and the kinds of relationships between the tasks based on them has received little attention. We explore a formalization of this question by considering a generative process where each task is associated with a high-dimensional manifold, which can be done in brain-like models with neuromodulatory systems. Following this formulation, we define Task-specific Geometric Sensitive Hashing and show that a randomly weighted neural network with a neuromodulation system can realize this function. 

Abstract Artificial neural networks are known to suffer from catastrophic forgetting: when learning multiple tasks sequentially, they perform well on the most recent task at the expense of previously learned tasks. In the brain, sleep is known to play an important role in incremental learning by replaying recent and old conflicting memory traces. Here we tested the hypothesis that implementing a sleep-like phase in artificial neural networks can protect old memories during new training and alleviate catastrophic forgetting. Sleep was implemented as off-line training with local unsupervised Hebbian plasticity rules and noisy input. In an incremental learning framework, sleep was able to recover old tasks that were otherwise forgotten. Previously learned memories were replayed spontaneously during sleep, forming unique representations for each class of inputs. Representational sparseness and neuronal activity corresponding to the old tasks increased while new task related activity decreased. The study suggests that spontaneous replay simulating sleep-like dynamics can alleviate catastrophic forgetting in artificial neural networks.