Abstract: Features related to systems and methods for automated generation of a machine learning model based in part on a pretrained model are described. The pretrained model is used as a starting point to augment and retrain according to client specifications. The identification of an appropriate pretrained model is based on the client specifications such as model inputs, model outputs, and similarities between the data used to train the models.

Abstract: Techniques are described automatically determining runtime configurations used to execute recurrent neural networks (RNNs) for training or inference. One such configuration involves determining whether to execute an RNN in a looped, or “rolled,” execution pattern or in a non-looped, or “unrolled,” execution pattern. Execution of an RNN using a rolled execution pattern generally consumes less memory resources than execution using an unrolled execution pattern, whereas execution of an RNN using an unrolled execution pattern typically executes faster. The configuration choice thus involves a time-memory tradeoff that can significantly affect the performance of the RNN execution. This determination is made automatically by a machine learning (ML) runtime by analyzing various factors such as, for example, a type of RNN being executed, the network structure of the RNN, characteristics of the input data to the RNN, an amount of computing resources available, and so forth.

Abstract: Features related to systems and methods for automated generation of a machine learning model based in part on a pretrained model are described. The pretrained model is used as a starting point to augment and retrain according to client specifications. The identification of an appropriate pretrained model is based on the client specifications such as model inputs, model outputs, and similarities between the data used to train the models.

Abstract: Methods and apparatus for dynamically reducing bandwidth usage by embedded displays are disclosed. An example method includes receiving a request to display a frame associated with a pixel depth on a display of a computing device, determining whether the frame includes a background layer, and when the frame includes the background layer, adjusting the pixel depth of the background layer.

Abstract: Techniques related to filtering hot plug signals are described herein. The techniques include receiving a first hot plug detect (HPD) signals and a second HPD signal from an external display device. A time period between receiving the first and second HPD signals is determined, and the first and second HPD signals are filtered based on the determined time period.

Abstract: Methods and apparatus for dynamically reducing bandwidth usage by embedded displays are disclosed. An example method includes receiving a request to display a frame associated with a pixel depth on a display of a computing device, determining whether the frame includes a background layer, and when the frame includes the background layer, adjusting the pixel depth of the background layer.

Abstract: Techniques related to filtering hot plug signals are described herein. The techniques include receiving a first hot plug detect (HPD) signals and a second HPD signal from an external display device. A time period between receiving the first and second HPD signals is determined, and the first and second HPD signals are filtered based on the determined time period.