{"url":"/method/causal-convolution","slug":"causal-convolution","name":"Causal Convolution","full_name":"Causal Convolution","full_name_withheld":false,"description_markdown":"**Causal convolutions** are a type of [convolution](https://paperswithcode.com/method/convolution) used for temporal data which ensures the model cannot violate the ordering in which we model the data: the prediction $p(x_{t+1} | x_{1}, \\ldots, x_{t})$ emitted by the model at timestep $t$ cannot depend on any of the future timesteps $x_{t+1}, x_{t+2}, \\ldots, x_{T}$. For images, the equivalent of a causal convolution is a [masked convolution](https://paperswithcode.com/method/masked-convolution) which can be implemented by constructing a mask tensor and doing an element-wise multiplication of this mask with the convolution kernel before applying it. For 1-D data such as audio one can more easily implement this by shifting the output of a normal convolution by a few timesteps.","description_state":"present","introduced_year":null,"introduced_by":{"title":"WaveNet: A Generative Model for Raw Audio","paper":"/paper/wavenet-a-generative-model-for-raw-audio","first_author":"Aaron van den Oord","n_authors":9,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/wavenet-a-generative-model-for-raw-audio"},"source":{"url":"http://arxiv.org/abs/1609.03499v2","title":"WaveNet: A Generative Model for Raw Audio","url_on_a_paper_host":true},"code_snippet_url":null,"code_snippet_url_on_a_code_host":false,"categories":[{"area":"Sequential","area_id":"sequential","collection":"Temporal Convolutions","url":"/methods/category/temporal-convolutions","pwc_aliases":[]}],"n_papers_tagged":27,"archive_num_papers":27,"papers_newest_first":[{"paper":null,"title":"Rolling Forward: Enhancing LightGCN with Causal Graph Convolution for Credit Bond Recommendation","date":"2025-03-18","arxiv_id":"2503.14213","n_code_links":0,"syntology":null},{"paper":null,"title":"Improved Video VAE for Latent Video Diffusion Model","date":"2024-11-10","arxiv_id":"2411.06449","n_code_links":0,"syntology":null},{"paper":"/paper/causal-discovery-from-time-series-data-with","title":"Causal Discovery from Time-Series Data with Short-Term Invariance-Based Convolutional Neural Networks","date":"2024-08-15","arxiv_id":"2408.08023","n_code_links":1,"syntology":null},{"paper":"/paper/causalformer-an-interpretable-transformer-for","title":"CausalFormer: An Interpretable Transformer for Temporal Causal Discovery","date":"2024-06-24","arxiv_id":"2406.16708","n_code_links":1,"syntology":{"ran":7,"of":7,"unverified":0,"pointer_only":7}},{"paper":"/paper/acco-automated-causal-cnn-scheduling","title":"ACCO: Automated Causal CNN Scheduling Optimizer for Real-Time Edge Accelerators","date":"2024-06-11","arxiv_id":"2406.07161","n_code_links":1,"syntology":null},{"paper":"/paper/mambats-improved-selective-state-space-models","title":"MambaTS: Improved Selective State Space Models for Long-term Time Series Forecasting","date":"2024-05-26","arxiv_id":"2405.16440","n_code_links":1,"syntology":{"ran":5,"of":7,"unverified":2,"pointer_only":0}},{"paper":null,"title":"Localization of Dummy Data Injection Attacks in Power Systems Considering Incomplete Topological Information: A Spatio-Temporal Graph Wavelet Convolutional Neural Network Approach","date":"2024-01-27","arxiv_id":"2401.15321","n_code_links":0,"syntology":null},{"paper":"/paper/ultra-short-term-multi-step-wind-speed","title":"Ultra-short-term multi-step wind speed prediction for wind farms based on adaptive noise reduction technology and temporal convolutional network","date":"2023-11-27","arxiv_id":"2311.16198","n_code_links":1,"syntology":null},{"paper":null,"title":"DualVC 2: Dynamic Masked Convolution for Unified Streaming and Non-Streaming Voice Conversion","date":"2023-09-27","arxiv_id":"2309.15496","n_code_links":0,"syntology":null},{"paper":"/paper/learning-minimal-representations-of","title":"Learning minimal representations of stochastic processes with variational autoencoders","date":"2023-07-21","arxiv_id":"2307.11608","n_code_links":1,"syntology":null},{"paper":"/paper/sequence-modeling-with-multiresolution","title":"Sequence Modeling with Multiresolution Convolutional Memory","date":"2023-05-02","arxiv_id":"2305.01638","n_code_links":1,"syntology":{"ran":3,"of":8,"unverified":5,"pointer_only":0}},{"paper":null,"title":"Dynamic Chunk Convolution for Unified Streaming and Non-Streaming Conformer ASR","date":"2023-04-18","arxiv_id":"2304.09325","n_code_links":0,"syntology":null},{"paper":"/paper/hgwavenet-a-hyperbolic-graph-neural-network","title":"HGWaveNet: A Hyperbolic Graph Neural Network for Temporal Link Prediction","date":"2023-04-14","arxiv_id":"2304.07302","n_code_links":1,"syntology":null},{"paper":"/paper/real-time-target-sound-extraction","title":"Real-Time Target Sound Extraction","date":"2022-11-04","arxiv_id":"2211.02250","n_code_links":1,"syntology":{"ran":3,"of":7,"unverified":4,"pointer_only":0}},{"paper":"/paper/towards-fast-single-trial-online-erp-based","title":"Towards Fast Single-Trial Online ERP based Brain-Computer Interface using dry EEG electrodes and neural networks: a pilot study","date":"2022-11-04","arxiv_id":"2211.10352","n_code_links":2,"syntology":null},{"paper":"/paper/gm-tcnet-gated-multi-scale-temporal","title":"GM-TCNet: Gated Multi-scale Temporal Convolutional Network using Emotion Causality for Speech Emotion Recognition","date":"2022-10-28","arxiv_id":"2210.15834","n_code_links":1,"syntology":null},{"paper":null,"title":"Locality-aware Attention Network with Discriminative Dynamics Learning for Weakly Supervised Anomaly Detection","date":"2022-08-11","arxiv_id":"2208.05636","n_code_links":0,"syntology":null},{"paper":"/paper/harmof0-logarithmic-scale-dilated-convolution","title":"HarmoF0: Logarithmic Scale Dilated Convolution For Pitch Estimation","date":"2022-05-02","arxiv_id":"2205.01019","n_code_links":1,"syntology":null},{"paper":null,"title":"PGCN: Progressive Graph Convolutional Networks for Spatial-Temporal Traffic Forecasting","date":"2022-02-18","arxiv_id":"2202.08982","n_code_links":0,"syntology":null},{"paper":null,"title":"Streaming Transformer Transducer Based Speech Recognition Using Non-Causal Convolution","date":"2021-10-07","arxiv_id":"2110.05241","n_code_links":0,"syntology":null},{"paper":"/paper/time-series-is-a-special-sequence-forecasting","title":"SCINet: Time Series Modeling and Forecasting with Sample Convolution and Interaction","date":"2021-06-17","arxiv_id":"2106.09305","n_code_links":6,"syntology":{"ran":1,"of":6,"unverified":5,"pointer_only":0}},{"paper":"/paper/transformer-transducer-end-to-end-speech","title":"Transformer-Transducer: End-to-End Speech Recognition with Self-Attention","date":"2019-10-28","arxiv_id":"1910.12977","n_code_links":1,"syntology":null},{"paper":null,"title":"MC\\^2: Multi-perspective Convolutional Cube for Conversational Machine Reading Comprehension","date":"2019-07-01","arxiv_id":null,"n_code_links":0,"syntology":null},{"paper":"/paper/enhancing-the-locality-and-breaking-the","title":"Enhancing the Locality and Breaking the Memory Bottleneck of Transformer on Time Series Forecasting","date":"2019-06-29","arxiv_id":"1907.00235","n_code_links":2,"syntology":null},{"paper":null,"title":"ConvS2S-VC: Fully convolutional sequence-to-sequence voice conversion","date":"2018-11-05","arxiv_id":"1811.01609","n_code_links":0,"syntology":null},{"paper":null,"title":"SeriesNet:A Generative Time Series Forecasting Model","date":"2018-08-23","arxiv_id":null,"n_code_links":0,"syntology":null},{"paper":"/paper/wavenet-a-generative-model-for-raw-audio","title":"WaveNet: A Generative Model for Raw Audio","date":"2016-09-12","arxiv_id":"1609.03499","n_code_links":62,"syntology":{"ran":41,"of":103,"unverified":62,"pointer_only":25}}],"papers_shown":27,"tasks":[{"task":"/task/time-series-1","name":"Time Series","papers":6},{"task":"/task/time-series","name":"Time Series Analysis","papers":4},{"task":"/task/time-series-forecasting","name":"Time Series Forecasting","papers":4},{"task":"/task/speech-recognition","name":"Speech Recognition","papers":3},{"task":"/task/model","name":"model","papers":3},{"task":"/task/speech-recognition-1","name":"speech-recognition","papers":3},{"task":"/task/causal-discovery","name":"Causal Discovery","papers":2},{"task":"/task/decoder","name":"Decoder","papers":2},{"task":"/task/representation-learning","name":"Representation Learning","papers":2},{"task":"/task/traffic-prediction","name":"Traffic Prediction","papers":2},{"task":"/task/voice-conversion","name":"Voice Conversion","papers":2},{"task":"/task/anomaly-detection","name":"Anomaly Detection","papers":1},{"task":"/task/audio-generation","name":"Audio Generation","papers":1},{"task":"/task/brain-computer-interface","name":"Brain Computer Interface","papers":1},{"task":"/task/conversational-question-answering","name":"Conversational Question Answering","papers":1},{"task":"/task/denoising","name":"Denoising","papers":1},{"task":"/task/density-estimation","name":"Density Estimation","papers":1},{"task":"/task/eeg-1","name":"EEG","papers":1},{"task":"/task/erp","name":"ERP","papers":1},{"task":"/task/edge-computing","name":"Edge-computing","papers":1}],"tasks_shown":20,"n_tasks":51,"usage_by_year":[{"year":"2016","papers":1},{"year":"2018","papers":2},{"year":"2019","papers":3},{"year":"2021","papers":2},{"year":"2022","papers":6},{"year":"2023","papers":6},{"year":"2024","papers":6},{"year":"2025","papers":1}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/causal-convolution"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}