Time2Vec: Learning a Vector Representation of Time

Abstract

Time is an important feature in many applications involving events that occur synchronously and/or asynchronously. To effectively consume time information, recent studies have focused on designing new architectures. In this paper, we take an orthogonal but complementary approach by providing a model-agnostic vector representation for time, called Time2Vec, that can be easily imported into many existing and future architectures and improve their performances. We show on a range of models and problems that replacing the notion of time with its Time2Vec representation improves the performance of the final model.

References in corpus (13)

• Adam: A Method for Stochastic Optimization
• Distributed Representations of Words and Phrases and their Compositionality
• LSTM: A Search Space Odyssey
• Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
• Convolutional Sequence to Sequence Learning
• A Review of Relational Machine Learning for Knowledge Graphs
• RetainVis: Visual Analytics with Interpretable and Interactive Recurrent Neural Networks on Electronic Medical Records
• Deep Reinforcement Learning of Marked Temporal Point Processes
• Neural Decomposition of Time-Series Data for Effective Generalization
• Learning Temporal Point Processes via Reinforcement Learning
• Streaming Graph Neural Networks
• Learning Dynamic Embeddings from Temporal Interactions
• evt_MNIST: A spike based version of traditional MNIST

Cited by in corpus (20)

• Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains
• Foundations and modelling of dynamic networks using Dynamic Graph Neural Networks: A survey
• A Review of Location Encoding for GeoAI: Methods and Applications
• Temporal Graph Networks for Deep Learning on Dynamic Graphs
• GC-LSTM: Graph Convolution Embedded LSTM for Dynamic Link Prediction
• Long-Range Transformers for Dynamic Spatiotemporal Forecasting
• Representation Learning for Dynamic Graphs: A Survey
• Inductive Representation Learning in Temporal Networks via Causal Anonymous Walks
• Network representation learning systematic review: ancestors and current development state
• Multi-Time Attention Networks for Irregularly Sampled Time Series
• Stochastic Latent Transformer: Efficient Modelling of Stochastically Forced Zonal Jets
• Training for the Future: A Simple Gradient Interpolation Loss to Generalize Along Time
• CRNNs for Urban Sound Tagging with spatiotemporal context
• Predicting Human Strategies in Simulated Search and Rescue Task
• A Deep Learning Framework for COVID Outbreak Prediction
• Individual Mobility Prediction via Attentive Marked Temporal Point Processes
• Topic-time Heatmaps for Human-in-the-loop Topic Detection and Tracking
• Narrative Cartography with Knowledge Graphs
• Towards Generating Real-World Time Series Data
• Inductive Representation Learning in Temporal Networks via Mining Neighborhood and Community Influences