• Tidak ada hasil yang ditemukan

Space Meets Time: Local Spacetime Neural Network For Traffic Flow Forecasting

Introduction

Chapter 3. Space Meets Time: Local Spacetime Neural Network For Traffic Flow Forecasting

In this chapter, we examine the traffic prediction problem and introduce aLocal Space-Time Neural Network (STNN). This innovative approach utilizes space- time convolution and attention mechanisms to learn the universal spatio-temporal correlations, effectively addressingRQ 1.

Thesis Outline 21

The growing influence of data-driven technologies in modern transportation systems has led to an increased focus on traffic flow forecasting. Accurate and timely predictions of traffic dynamics can significantly enhance transportation management, alleviate congestion, and improve overall efficiency. Traffic systems, characterized by changing flows in road networks, exhibit salient patterns influ- enced by various extrinsic factors and intrinsic principles. Accurate traffic flow predictions rely on a model’s ability to capture not only extrinsic features but also the intrinsic, universal patterns that govern traffic flow. Uncovering these patterns and understanding the latent correlations between a location’s current state and its surrounding locations’ past are crucial for developing effective traffic forecasting models.

Recent advancements in neural-based techniques, especially GNNs, have sig- nificantly improved traffic feature representation and prediction results compared to earlier statistical methods. However, existing GNN-based traffic forecasting models face three major challenges: (1) their heavy reliance on graph structure limits their applicability to specific road networks, preventing the discovery of intrinsic traffic system properties; (2) the computationally expensive feature aggre- gation operations, such as graph convolutions, impede scalability for large road networks with numerous sensors; and (3) the separate components for spatial and temporal feature extraction in these models assume uniform correlations between locations over time, which may not accurately reflect the dynamic nature of traffic systems.

In order to solve the above challenges andRQ 1, we begin by defining several key concepts that will be useful throughout the thesis. One such concept is the traffic event, which integrates both the spatial and temporal aspects of a traffic measurement collected by a sensor station. Drawing a parallel to the concept of events in physics [22], this concept will prove essential in our efforts to solve the challenges at hand.

Definition 1(Traffic Event). Given a traffic measurement s (e.g., speed) observed at sensor vi and time t, a traffic event is a tuple consists of the measurement, time, and location, namely,(s,t,vi).

Accordingly,we can generalize this definition to encompass a broader range of scenarios, not solely limited to traffic prediction.

22 Introduction

Definition 2(Spacetime Event). Given any sensor measurement m observed at location viand time t, a spacetime event is a tuple consists of the measurement, time, and location, namely,(m,t,vi).

Additionally, the essential concept for this chapter, as well as the entire thesis, is thespacetime interval. Thespacetime intervalbetween two traffic events signifies the degree to which one event impacts the other; a smaller interval indicates a stronger connection between the two traffic events. Within a local-spacetime context, we are primarily concerned with the intervals between traffic events at the target sensor and those at other sensors.

Definition 3(Spacetime Interval). Spacetime interval is the quantified influence of a traffic event imposed on another traffic event regarding to the traffic measurement.

Figure 3.1 provides an illustration of the spacetime interval concept. The figure displays three instances of the network. The state of the target node att3is significantly affected by the state ofaatt1and the state ofbatt2. However, it is only mildly influenced by the state ofaatt2and the state ofbatt1.

time

target spacetime interval

1

0

impact

a

b

a a

b b

t1 t2 t3

Figure 3.1: An illustration of spacetime interval

Armed with these key concepts, we propose a novel spatio-temporal correla- tion learning paradigm called Spacetime Interval Learning. This approach fuses spatial and temporal dimensions into a single manifold, referred to as space- time, and captures correlations as intervals between traffic events. The paradigm extracts traffic data from nearby sensors within a fixed time window, called the local-spacetime context, which allows the model to focus on relevant sen- sors. Our method correlates nodes at different times within the local-spacetime context, resulting in a model that is universal, independent of graph structure,

Thesis Outline 23

and applicable to various traffic systems. By shifting the focus from network- level to node-level predictions, our approach facilitates parallel predictions for multiple locations and efficiently captures varying spatial correlations between locations over time. The architecture of the model is shown in Figure 3.2. The STNN compriseskspacetime modules (ST-Modules) and a fully-connected out- put layer. Each ST-Module contains a spacetime attention block (ST-Attn block) and a spacetime convolution block (ST-Conv block). The ST-Attn block utilizes a self-attention mechanism to emphasize the most influential traffic events. Within each ST-Conv block, three distinct convolution kernels are employed to aggregate the spatio-temporal correlations from various perspectives. Subsequently, the extracted features are stacked and condensed using a 1×1 convolution.

time

Sub-spacetime Input network

ST-Conv

Input:

Output:

...

ST-Atten

Residual Connections

1x1 Conv

ST-Conv Block ST-Attn Block

ST-Conv ST-Atten

1x1 Conv FC

Input: Output:

Residual Connections Residual Connections

ST-Conv ST-Attn

Residual Connections ST-Conv Block ST-Attn Block

ST ModuleST Module

Linear

Conv Conv

condense ST-Conv Block

concat Conv

...

...

Figure 3.2: The architecture of STNN with an example local-spacetime.

At last, the proposed STNN model was evaluated in various settings to demon- strate its robustness and ability to generalize to unseen traffic networks, unlike existing state-of-the-art models. In the first setting, STNN was trained and tested on the same network, where it surpassed baselines in terms of prediction accu- racy. In the second setting, STNN was trained on one network and tested on a different one, and its performance was found to be comparable to state-of-the- art methods trained directly on the test dataset. In the third setting, STNN was trained on multiple networks and tested on each, yielding performance close to models trained and tested on the same network while outperforming baselines on short-term predictions. When evaluated on a simulated dynamic network, STNN outperformed baselines by a large margin, showcasing its ability to adapt to changing network topologies.

In conclusion, the contribution of the proposed STNN model is that it offers accuracy, transferrability, the ability to handle dynamic networks, and scalability.

24 Introduction

The STNN model, which learns spacetime intervals, effectively captures correla- tions between locations across different time instances, leading to improved traffic predictions. Experiments on two real-world datasets, METR-LA and PeMS-Bay, show that the model outperforms existing methods, with notable improvements in accuracy. The model’s transferrability is demonstrated through its performance on unseen traffic datasets without fine-tuning. Additionally, the STNN can handle dynamic networks, such as networks with changing topologies, and its scalability allows for application on networks of arbitrary size, unlike previous GNN-based models. Therefore, theRQ 1is solved smoothly in this chapter.

Chapter 4. GETNext: Trajectory Flow Map Enhanced Transformer for