Therefore, from a system-level perspective, the design ethos of a system-efficient communication-compression algorithm is that we should guarantee that the compression/decompression of the algorithm is computationally light and takes less time, and it should also be friendly to efficient collective communication primitives.

Various gradient compression algorithms have been proposed to alleviate the communication bottleneck in distributed learning, and they have demonstrated effectiveness in terms of high compression ratios and theoretical low communication complexity. However, when it comes to practically training modern deep neural networks (DNNs), these algorithms have yet to match the inference performance of uncompressed SGD-momentum (SGDM) and adaptive optimizers (e.g.,Adam). More importantly, recent studies suggest that these algorithms actually offer no speed advantages over SGDM/Adam when used with common distributed DNN training frameworks ( e.g., DistributedDataParallel (DDP)) in the typical settings, due to heavy compression/decompression computation or incompatibility with the efficient All-Reduce or the requirement of uncompressed warmup at the early stage. For these reasons, we propose a novel 1-bit adaptive optimizer, dubbed *Bi*nary *r*andomization a*d*aptive optimiz*er* (**Birder**). The quantization of Birder can be easily and lightly computed, and it does not require warmup with its uncompressed version in the beginning. Also, we devise Hierarchical-1-bit-All-Reduce to further lower the communication volume. We theoretically prove that it promises the same convergence rate as the Adam. Extensive experiments, conducted on 8 to 64 GPUs (1 to 8 nodes) using DDP, demonstrate that Birder achieves comparable inference performance to uncompressed SGDM/Adam, with up to ${2.5 \times}$ speedup for training ResNet-50 and ${6.3\times}$ speedup for training BERT-Base. Code is publicly available at https://openi.pcl.ac.cn/c2net_optim/Birder.

Therefore, from a system-level perspective, the design ethos of a system-efficient communication-compression algorithm is that we should guarantee that the compression/decompression of the algorithm is computationally light and takes less time, and it should also be friendly to efficient collective communication primitives.

Various gradient compression algorithms have been proposed to alleviate the communication bottleneck in distributed learning, and they have demonstrated effectiveness in terms of high compression ratios and theoretical low communication complexity. However, when it comes to practically training modern deep neural networks (DNNs), these algorithms have yet to match the inference performance of uncompressed SGD-momentum (SGDM) and adaptive optimizers (e.g.,Adam). More importantly, recent studies suggest that these algorithms actually offer no speed advantages over SGDM/Adam when used with common distributed DNN training frameworks ( e.g., DistributedDataParallel (DDP)) in the typical settings, due to heavy compression/decompression computation or incompatibility with the efficient All-Reduce or the requirement of uncompressed warmup at the early stage. For these reasons, we propose a novel 1-bit adaptive optimizer, dubbed *Bi*nary *r*andomization a*d*aptive optimiz*er* (**Birder**). The quantization of Birder can be easily and lightly computed, and it does not require warmup with its uncompressed version in the beginning.

An immense frustration ecologists encounter is prompted by the attempt to keep track of individual animals in a study. This task only becomes more difficult when trying to pinpoint small, mobile animals like songbirds. While intelligent computer algorithms can help scientists better complete this task, training these systems to recognize different species -- let alone individuals in a species -- can take thousands of data points, time, and money. However, French and Portuguese researchers recently devised a way to streamline this process. They designed a deep-learning network that can identify individual birds with up to 92 percent accuracy in three different species. This tech can not only save scientists resources but can help them collect important data about the lives of birds -- and better understand what may be leading to their decline in North America.

I can't count the number of animal stories that appeared in my timelines this year with comments like, "Everything is garbage, so here's this." There was the cat who was reunited with her family after the Camp Fire, in California, and the parrot who was adopted after getting kicked out of an animal shelter for swearing too saltily. Among the bears preparing for hibernation at Katmai National Park, a female named Beadnose became famous for being the most gloriously round. There was the baby raccoon who scaled a skyscraper in St. Paul, "Mission Impossible" style, stopping occasionally for naps in window ledges along the way. Stories from the animal world offer reliable moments of escapism--the ones we see in viral videos are usually cute, or tame, or strange and majestic, and glimpsed from a safe distance.

Bird populations are plummeting, thanks to logging, agriculture, and climate change. Scientists keep track of species by recording their calls, but even the best computer programs can't reliably distinguish bird calls from other sounds. Now, thanks to a bit of crowdsourcing and a lot of artificial intelligence (AI), researchers say they have something to crow about. AI algorithms can be as finicky as finches, often requiring manual calibration and retraining for each new location or species. So an interdisciplinary group of researchers launched the Bird Audio Detection challenge, which released hours of audio from environmental monitoring stations around Chernobyl, Ukraine, which they happened to have access to, as well as crowdsourced recordings, some of which came from an app called Warblr.