From smartphone assistants to image recognition and translation, machine learning already helps us in our everyday lives. But it can also help us to tackle some of the world's most challenging physical problems -- such as energy consumption. Large-scale commercial and industrial systems like data centres consume a lot of energy, and while much has been done to stem the growth of energy use, there remains a lot more to do given the world's increasing need for computing power. Reducing energy usage has been a major focus for us over the past 10 years: we have built our own super-efficient servers at Google, invented more efficient ways to cool our data centres and invested heavily in green energy sources, with the goal of being powered 100 percent by renewable energy. Compared to five years ago, we now get around 3.5 times the computing power out of the same amount of energy, and we continue to make many improvements each year.
As created for AI4IMPACT's Deep Learning Datathon 2020, TEAM DEFAULT has created a neural-network-based deep learning model used for predicting energy production demand in France. The model was created using Smojo, on AI4IMPACT's innovative cloud-based learning and model deployment system. Our model was able to achieve a 0.131 test loss which beat persistence loss of 0.485 by a quite a fair margin. As the energy market becomes increasingly liberalized across the world, the free and open market has seen an uptick and importance for optimized energy demand. New and existing entrants turn to data and various methods to forecast energy consumption in hopes of turning over a profit.
Convolutional neural networks (CNNs) have been increasingly deployed to edge devices. Hence, many efforts have been made towards efficient CNN inference on resource-constrained platforms. This paper attempts to explore an orthogonal direction: how to conduct more energy-efficient training of CNNs, so as to enable on-device training? We strive to reduce the energy cost during training, by dropping unnecessary computations, from three complementary levels: stochastic mini-batch dropping on the data level; selective layer update on the model level; and sign prediction for low-cost, low-precision back-propagation, on the algorithm level. Extensive simulations and ablation studies, with real energy measurements from an FPGA board, confirm the superiority of our proposed strategies and demonstrate remarkable energy savings for training. For example, when training ResNet-74 on CIFAR-10, we achieve aggressive energy savings of >90% and >60%, while incurring a top-1 accuracy loss of only about 2% and 1.2%, respectively. When training ResNet-110 on CIFAR-100, an over 84% training energy saving is achieved without degrading inference accuracy.
Carbon-free technologies like renewable energy help combat climate change, but many of them have not reached their full potential. Consider wind power: over the past decade, wind farms have become an important source of carbon-free electricity as the cost of turbines has plummeted and adoption has surged. However, the variable nature of wind itself makes it an unpredictable energy source--less useful than one that can reliably deliver power at a set time. In search of a solution to this problem, last year, DeepMind and Google started applying machine learning algorithms to 700 megawatts of wind power capacity in the central United States. These wind farms--part of Google's global fleet of renewable energy projects--collectively generate as much electricity as is needed by a medium-sized city.
The second annual conference, organized by the conservative thinktank the Heartland Institute, pulled together speakers from JunkScience, the Committee for a Constructive Tomorrow, and the Center For Industrial Progress, along with officials from the U.S. Department of Interior and the White House for panels that included: "Carbon Taxes, Cap & Trade, and Other Bad Ideas," "Fiduciary Malpractice: The Sustainable Investment Movement," and "Why CO2 Emissions Are Not Creating A Climate Crisis."