-
The Journal of Financial Data Science Summer2021The authors enhance pretrained language models with Securities and Exchange Commission filings data to create better language representations for features used in a predictive model. Specifically, they train RoBERTa class models with additional financial regulatory text, which they denote as a class of RoBERTa-Fin models. Using different datasets, the authors assess whether there is material improvement
-
Journal of Political Economy2021Calibration means that forecasts and average realized frequencies are close. We develop the concept of forecast hedging, which consists of choosing the forecasts so as to guarantee that the expected track record can only improve. This yields all the calibration results by the same simple basic argument while differentiating between them by the forecast-hedging tools used: deterministic and fixed point based
-
COLING 20202020End-to-end neural data-to-text (D2T) generation has recently emerged as an alternative to pipeline-based architectures. However, it has faced challenges generalizing to new domains and generating semantically consistent text. In this work, we present DATATUNER, a neural, end-to-end data-to-text generation system that makes minimal assumptions about the data representation and target domain. We take a two-stage
-
ICML 20202020We propose a reward function estimation framework for inverse reinforcement learning with deep energy-based policies. We name our method PQR, as it sequentially estimates the Policy, the Qfunction, and the Reward function. PQR does not assume that the reward solely depends on the state, instead it allows for a dependency on the choice of action. Moreover, PQR allows for stochastic state transitions. To
-
AEA 20202020Inflation indices are important inputs into measuring aggregate productivity and cost of living and conducting monetary and economic policy. We want to contribute to the science of inflation measurement based on quality-adjusted (hedonic) prices. The main challenges today are: (1) the global trade environment (millions of products); (2) frequent price changes; (3) extremely high turnover for some products
Related content
-
April 26, 2021New method identifies which causal factors contribute most to observed changes in probability distributions.
-
April 13, 2021Their doctoral degrees help these product managers bridge the gap between business and science.
-
February 18, 2021The Amazon research manager was included on a list of individuals 40 and younger who are projected to play a leading role in Africa’s economic future.
-
January 11, 2021Machine learning is helping Amazon improve the grocery shopping experience in both the physical and online channels.
-
January 6, 2021Amazon Scholar David Card on the revolution in economic research that he helped launch and its consequences for industry.
-
November 9, 2020Amazon VP and chief economist for digital streaming and advertising Phil Leslie on economists’ role in industry.