CAMul: Calibrated and Accurate Multi-view Time-Series Forecasting

  title={CAMul: Calibrated and Accurate Multi-view Time-Series Forecasting},
  author={Harshavardhan Kamarthi and Lingkai Kong and Alexander Rodr'iguez and Chao Zhang and B. Aditya Prakash},
  journal={Proceedings of the ACM Web Conference 2022},
Probabilistic time-series forecasting enables reliable decision making across many domains. Most forecasting problems have diverse sources of data containing multiple modalities and structures. Leveraging information from these data sources for accurate and well-calibrated forecasts is an important but challenging problem. Most previous works on multi-view time-series forecasting aggregate features from each data view by simple summation or concatenation and do not explicitly model uncertainty… 

Figures and Tables from this paper

Data-Centric Epidemic Forecasting: A Survey

This survey delves into various data-driven methodological and practical advancements and introduces a conceptual framework to navigate through them, enumerate the large number of epidemiological datasets and novel data streams that are relevant to epidemic forecasting, and highlights some challenges and open problems found across the forecasting pipeline.

Differentiable agent-based epidemiological modeling for end-to-end learning

G RAD ABM is introduced which is a new scalable, fast and differentiable design for ABMs that makes it amenable to be merged with deep neural networks and seamlessly integrate heterogeneous data sources to help with calibration, forecasting and policy evaluation.

Differentiable Agent-based Epidemiology

G RAD ABM is introduced which is a new scalable, fast and differentiable design for ABMs that makes it amenable to be merged with deep neural networks and seamlessly integrate heterogeneous data sources to help with calibration, forecasting and policy evaluation.

PROFHIT: Probabilistic Robust Forecasting for Hierarchical Time-series

This work proposes PROFH I T, which is a fully probabilistic hierarchical forecasting model that jointly models forecast distribution of entire hierarchy and introduces a novel Distributional Coherency regularization to learn from hierarchical relations for entire forecast distribution that enables robust and calibrated forecasts.

EINNs: Epidemiologically-Informed Neural Networks

This work introduces a new class of physics-informed neural networks-EINNcrafted for epidemic forecasting that allows neural models have the flexibility to learn the disease spread dynamics and use auxiliary features in a general framework.



Evaluation of individual and ensemble probabilistic forecasts of COVID-19 mortality in the US

This project systematically evaluates 23 models that regularly submitted forecasts of reported weekly incident COVID-19 mortality counts in the US at the state and national level and underscores the role that collaboration and active coordination between governmental public health agencies, academic modeling teams, and industry partners can play in developing modern modeling capabilities to support local, state, and federal response to outbreaks.

Recurrent Attentive Neural Process for Sequential Data

The proposed Recurrent Attentive Neural Process (RANP) encapsulates both the inductive biases of recurrent neural networks and also the strength of NPs for modelling uncertainty and can effectively model sequential data and outperforms NPs and LSTMs remarkably in a 1D regression toy example as well as autonomous-driving applications.

The Functional Neural Process

A new family of exchangeable stochastic processes, the Functional Neural Processes (FNPs), are presented and it is demonstrated that they are scalable to large datasets through mini-batch optimization and described how they can make predictions for new points via their posterior predictive distribution.

DeepCOVID: An Operational Deep Learning-driven Framework for Explainable Real-time COVID-19 Forecasting

DeepCOVID, an operational deep learning framework designed for real-time COVID-19 forecasting, works well with sparse data and can handle noisy heterogeneous data signals by propagating the uncertainty from the data in a principled manner resulting in meaningful uncertainties in the forecast.

Learning Interpretable Deep State Space Model for Probabilistic Time Series Forecasting

This work proposes a deep state space model for probabilistic time series forecasting whereby the non-linear emission model and transition model are parameterized by networks and the dependency is modeled by recurrent neural nets.

When in Doubt: Neural Non-Parametric Uncertainty Quantification for Epidemic Forecasting

This work model the forecasting task as a probabilistic generative process and proposes a functional neural process model called EPIFNP, which directly models the probability density of the forecast value and significantly outperforms previous state-of-the-art models in both accuracy and calibration metrics.

Graph Deep Factors for Forecasting with Applications to Cloud Resource Allocation

This work proposes a deep hybrid probabilistic graph-based forecasting framework called Graph Deep Factors (GraphDF), a hybrid forecasting framework that consists of a relational global and relational local model that learns complex non-linear time-series patterns globally using the structure of the graph to improve both forecasting accuracy and computational efficiency.

Back2Future: Leveraging Backfill Dynamics for Improving Real-time Predictions in Future

This paper construct a detailed dataset composed of relevant signals over the past year of the pandemic, and systematically characterize several patterns in backfill dynamics and leverage the observations for formulating a novel problem and neural framework Back2Future that aims to refines a given model's predictions in real-time.

DF-TAR: A Deep Fusion Network for Citywide Traffic Accident Risk Prediction with Dangerous Driving Behavior

This paper proposes a Deep Fusion network for citywide Traffic Accident Risk prediction (DF-TAR) with dangerous driving statistics that contain the frequencies of various dangerous driving offences in each region, and shows a strong correlation between them in terms of both location and time.

Wiki2Prop: A Multimodal Approach for Predicting Wikidata Properties from Wikipedia

This work proposes to identify and rank the properties that an entity might be missing in Wikidata, and solves the problem using a multimodal Deep Learning method to predict which properties should be attached to a given entity, using its Wikipedia page embeddings.