Amazon Technologies, Inc.

Sourav Kumar Agarwal

Amber Roy Chowdhury

Valerie Galluzzi Liptak

Tarun Bhatia

Satyam Saxena

Alok Chandra

Techniques are described for predicting a future defect and triggering a mitigation action. In an example, a system generates a first input to a causal model based on geographical data associated with a location and area data associated with one or more areas that include the location, and determines a first output of the causal model. The first output indicates a first prediction of a cause of a past defect for a service associated with the location. The system also generates a second input to a predictive model based on the first input and the first output, and determines a second output of the predictive model. The second output indicates a second prediction of a future defect associated with the service. Based on the second prediction, a mitigation action can be performed such that the future defect is prevented.

Triggering a service mitigation action based on a causal-predictive system

Generally described, one or more aspects of the present application correspond to a machine learning address normalization system. A system of deep learning networks can normalize the tokens of a free-form address into an address component hierarchy. Feature vectors representing various characters and words of the address tokens can be input into a bi-directional long short term memory network (LSTM) to generate a hidden state representation of each token, which can be individually passed through a softmax layer to generate probabilistic values of the token being each of the components in the address hierarchy. Thereafter, a conditional random field (CRF) model can select a particular address component for each token by using learned parameters to optimize a path through the collective outputs of the softmax layer for the tokens. Thus, the free-form address can be normalized to determine the values it contains for different components of a specified address hierarchy.

Growing community of inventors

Hyderabad, India

Sourav Kumar Agarwal