Modular approaches that use a different composition of modules for each problem are a promising direction in continual learning (CL). However, searching through the large, discrete space of module compositions is challenging, especially because evaluating a composition's performance requires a round of neural network training. We address this challenge through a modular CL framework, PICLE, that uses a probabilistic model to cheaply compute the fitness of each composition, allowing PICLE to achieve both perceptual, few-shot and latent transfer. The model combines prior knowledge about good module compositions with dataset-specific information. We evaluate PICLE using two benchmark suites designed to assess different desiderata of CL techniques. Comparing to a wide range of approaches, we show that PICLE is the first modular CL algorithm to achieve perceptual, few-shot and latent transfer while scaling well to large search spaces, outperforming previous state-of-the-art modular CL approaches on long problem sequences.

翻译：模块化方法通过针对每个问题使用不同的模块组合，是持续学习（CL）领域一个有前景的方向。然而，在庞大且离散的模块组合空间中进行搜索极具挑战，尤其因为评估组合性能需要一轮神经网络训练。我们通过一个名为PICLE的模块化CL框架解决了这一挑战，该框架利用概率模型低成本计算每种组合的适应度，使PICLE能够同时实现感知迁移、小样本迁移和潜在迁移。该模型将关于良好模块组合的先验知识与特定数据集的信息相结合。我们使用两套旨在评估CL技术不同需求基准测试套件对PICLE进行评价。与多种方法对比表明，PICLE是首个在扩展至大规模搜索空间的同时实现感知迁移、小样本迁移和潜在迁移的模块化CL算法，在长问题序列上优于先前最先进的模块化CL方法。