In autonomous driving, predicting the behavior (turning left, stopping, etc.)
of target vehicles is crucial for the self-driving vehicle to make safe
decisions and avoid accidents. Existing deep learning-based methods have shown
excellent and accurate performance, but the black-box nature makes it
untrustworthy to apply them in practical use. In this work, we explore the
interpretability of behavior prediction of target vehicles by an Episodic
Memory implanted Neural Decision Tree (abbrev. eMem-NDT). The structure of
eMem-NDT is constructed by hierarchically clustering the text embedding of
vehicle behavior descriptions. eMem-NDT is a neural-backed part of a
pre-trained deep learning model by changing the soft-max layer of the deep
model to eMem-NDT, for grouping and aligning the memory prototypes of the
historical vehicle behavior features in training data on a neural decision
tree. Each leaf node of eMem-NDT is modeled by a neural network for aligning
the behavior memory prototypes. By eMem-NDT, we infer each instance in behavior
prediction of vehicles by bottom-up Memory Prototype Matching (MPM) (searching
the appropriate leaf node and the links to the root node) and top-down Leaf
Link Aggregation (LLA) (obtaining the probability of future behaviors of
vehicles for certain instances). We validate eMem-NDT on BLVD and LOKI
datasets, and the results show that our model can obtain a superior performance
to other methods with clear explainability. The code is available at
this https URL

通过 Episodic Memory implanted Neural Decision Tree (eMem-NDT) 来解释目标车辆的行为预测，模型具有清晰的可解释性、优异的性能和适用于实际使用的特点。

通过嵌入情景记忆的 NDT 实现车辆行为预测

Vehicle Behavior Prediction by Episodic-Memory Implanted NDT

Though traditional algorithms could be embedded into neural architectures
with the proposed principle of \cite{xiao2017hungarian}, the variables that
only occur in the condition of branch could not be updated as a special case.
To tackle this issue, we multiply the conditioned branches with Dirac symbol
(i.e. $\mathbf{1}_{x>0}$), then approximate Dirac symbol with the continuous
functions (e.g. $1 - e^{-\alpha|x|}$). In this way, the gradients of
condition-specific variables could be worked out in the back-propagation
process, approximately, making a fully functioned neural graph. Within our
novel principle, we propose the neural decision tree \textbf{(NDT)}, which
takes simplified neural networks as decision function in each branch and
employs complex neural networks to generate the output in each leaf. Extensive
experiments verify our theoretical analysis and demonstrate the effectiveness
of our model.

通过提出的神经图结构来解决在分支条件中只出现变量无法特别更新的问题，并利用 Dirac 符号和连续函数来近似计算条件特定变量的梯度，最终提出了一种使用简化神经网络作为分支条件函数，复杂神经网络作为叶子输出的神经决策树模型，并经过大量实验证实该模型的有效性。