Department of Bioengineering, Graduate School of Engineering, The University of Tokyo, Tokyo, Japan.
Department of Research and Development, SoftBank Robotics Group Corp., Tokyo, Japan.
PLoS One. 2020 May 22;15(5):e0233559. doi: 10.1371/journal.pone.0233559. eCollection 2020.
Bayesian inference is the process of narrowing down the hypotheses (causes) to the one that best explains the observational data (effects). To accurately estimate a cause, a considerable amount of data is required to be observed for as long as possible. However, the object of inference is not always constant. In this case, a method such as exponential moving average (EMA) with a discounting rate is used to improve the ability to respond to a sudden change; it is also necessary to increase the discounting rate. That is, a trade-off is established in which the followability is improved by increasing the discounting rate, but the accuracy is reduced. Here, we propose an extended Bayesian inference (EBI), wherein human-like causal inference is incorporated. We show that both the learning and forgetting effects are introduced into Bayesian inference by incorporating the causal inference. We evaluate the estimation performance of the EBI through the learning task of a dynamically changing Gaussian mixture model. In the evaluation, the EBI performance is compared with those of the EMA and a sequential discounting expectation-maximization algorithm. The EBI was shown to modify the trade-off observed in the EMA.
贝叶斯推断是一种将假设（原因）缩小到最能解释观测数据（效果）的过程。为了准确估计原因，需要尽可能长时间地观察大量数据。然而，推断的对象并不总是不变的。在这种情况下，使用具有折扣率的指数移动平均 (EMA) 等方法来提高对突然变化的响应能力；还需要增加折扣率。也就是说，在提高折扣率以提高跟踪能力的同时，降低了准确性，从而建立了一种权衡。在这里，我们提出了一种扩展的贝叶斯推断 (EBI)，其中包含了类似于人类的因果推断。我们表明，通过将因果推断纳入贝叶斯推断，可以引入学习和遗忘效应。我们通过动态变化的高斯混合模型的学习任务来评估 EBI 的估计性能。在评估中，将 EBI 的性能与 EMA 和顺序折扣期望最大化算法的性能进行了比较。结果表明，EBI 可以修改 EMA 中观察到的权衡。
