Leveraging dendritic properties to advance machine learning and neuro-inspired computing.

The brain is a remarkably capable and efficient system. It can process and store huge amounts of noisy and unstructured information, using minimal energy. In contrast, current artificial intelligence (AI) systems require vast resources for training while still struggling to compete in tasks that are trivial for biological agents. Thus, brain-inspired engineering has emerged as a promising new avenue for designing sustainable, next-generation AI systems. Here, we describe how dendritic mechanisms of biological neurons have inspired innovative solutions for significant AI problems, including credit assignment in multi-layer networks, catastrophic forgetting, and high-power consumption. These findings provide exciting alternatives to existing architectures, showing how dendritic research can pave the way for building more powerful and energy efficient artificial learning systems.

A Generative Neighborhood-Based Deep Autoencoder for Robust Imbalanced Classification.

Deep learning models perform remarkably well on many classification tasks recently. The superior performance of deep neural networks relies on the large number of training data, which at the same time must have an equal class distribution in order to be efficient. However, in most real-world applications, the labeled data may be limited with high imbalance ratios among the classes, and thus, the learning process of most classification algorithms is adversely affected resulting in unstable predictions and low performance. Three main categories of approaches address the problem of imbalanced learning, i.e., data-level, algorithmic level, and hybrid methods, which combine the two aforementioned approaches. Data generative methods are typically based on generative adversarial networks, which require significant amounts of data, while model-level methods entail extensive domain expert knowledge to craft the learning objectives, thereby being less accessible for users without such knowledge. Moreover, the vast majority of these approaches are designed and applied to imaging applications, less to time series, and extremely rare to both of them. To address the above issues, we introduce GENDA, a generative neighborhood-based deep autoencoder, which is simple yet effective in its design and can be successfully applied to both image and time-series data. GENDA is based on learning latent representations that rely on the neighboring embedding space of the samples. Extensive experiments, conducted on a variety of widely-used real datasets demonstrate the efficacy of the proposed method. Impact Statement­: Imbalanced data classification is an actual and important issue in many real-world learning applications hampering most classification tasks. Fraud detection, biomedical imaging categorizing healthy people versus patients, and object detection are some indicative domains with an economic, social and technological impact, which are greatly affected by inherent imbalanced data distribution. However, the majority of the existing algorithms that address the imbalanced classification problem are designed with a particular application in mind, and thus they can be used with specific datasets and even hyperparameters. The generative model introduced in this paper overcomes this limitation and produces improved results for a large class of imaging and time series data even under severe imbalance ratios, making it quite competitive.

A dendritic substrate for temporal diversity of cortical inhibition.

In the mammalian neocortex, GABAergic interneurons (INs) inhibit cortical networks in profoundly different ways. The extent to which this depends on how different INs process excitatory signals along their dendrites is poorly understood. Here, we reveal that the functional specialization of two major populations of cortical INs is determined by the unique association of different dendritic integration modes with distinct synaptic organization motifs. We found that somatostatin (SST)-INs exhibit NMDAR-dependent dendritic integration and uniform synapse density along the dendritic tree. In contrast, dendrites of parvalbumin (PV)-INs exhibit passive synaptic integration coupled with proximally enriched synaptic distributions. Theoretical analysis shows that these two dendritic configurations result in different strategies to optimize synaptic efficacy in thin dendritic structures. Yet, the two configurations lead to distinct temporal engagement of each IN during network activity. We confirmed these predictions with in vivo recordings of IN activity in the visual cortex of awake mice, revealing a rapid and linear recruitment of PV-INs as opposed to a long-lasting integrative activation of SST-INs. Our work reveals the existence of distinct dendritic strategies that confer distinct temporal representations for the two major classes of neocortical INs and thus dynamics of inhibition.