In this paper, we introduce a new channel pruning method to accelerate very deep convolutional neural networks. Given a trained CNN model, we propose an iterative two-step algorithm to effectively prune each layer, by a LASSO regression based channel selection and least square reconstruction. We further generalize this algorithm to multi-layer and multi-branch cases. Our method reduces the accumulated error and enhance the compatibility with various architectures. Our pruned VGG-16 achieves the state-of-the-art results by 5× speed-up along with only 0.3% increase of error. More importantly, our method is able to accelerate modern networks like ResNet, Xception and suffers only 1.4%, 1.0% accuracy loss under 2× speedup respectively, which is significant. Code has been made publicly available 1 .

Model compression is a critical technique to efficiently deploy neural network models on mobile devices which have limited computation resources and tight power budgets. Conventional model compression techniques rely on hand-crafted heuristics and rule-based policies that require domain experts to explore the large design space trading off among model size, speed, and accuracy, which is usually sub-optimal and time-consuming. In this paper, we propose AutoML for Model Compression (AMC) which leverage reinforcement learning to provide the model compression policy. This learning-based compression policy outperforms conventional rule-based compression policy by having higher compression ratio, better preserving the accuracy and freeing human labor. Under 4× FLOPs reduction, we achieved 2.7% better accuracy than the handcrafted model compression policy for VGG-16 on ImageNet. We applied this automated, push-the-button compression pipeline to MobileNet and achieved 1.81× speedup of measured inference latency on an Android phone and 1.43× speedup on the Titan XP GPU, with only 0.1% loss of ImageNet Top-1 accuracy. Reward= -Error*log(FLOP) Agent: DDPG Action: Compress with Sparsity ratio at (e.g. 50%) Embedding st=[N,C,H,W,i…] Environment: Channel Pruning Layer t-1 Layer t Layer t+1 Critic Actor Embedding Original NN Model Compression by Human: Labor Consuming, Sub-optimal Model Compression by AI: Automated, Higher Compression Rate, Faster Compressed NN AMC Engine Original NN Compressed NN 30% 50% ? %

Gamma-ray detection and spectroscopy is the quantitative determination of their energy spectra, and is of critical value and critically important in diverse technological and scientific fields. Here we report an improved melt growth method for cesium lead bromide and a special detector design with asymmetrical metal electrode configuration that leads to a high performance at room temperature. As-grown centimeter-sized crystals possess extremely low impurity levels (below 10 p.p.m. for total 69 elements) and detectors achieve 3.9% energy resolution for 122 keV 57Co gamma-ray and 3.8% for 662 keV 137Cs gamma-ray. Cesium lead bromide is unique among all gamma-ray detection materials in that its hole transport properties are responsible for the high performance. The superior mobility-lifetime product for holes (1.34 × 10−3 cm2 V−1) derives mainly from the record long hole carrier lifetime (over 25 μs). The easily scalable crystal growth and high-energy resolution, highlight cesium lead bromide as an exceptional next generation material for room temperature radiation detection.

The organic-inorganic hybrid lead halide perovskites have emerged as a series of star materials for solar cells, lasers and detectors. However, the issues raised by the toxic lead element and marginal stability due to the volatile organic components have severely limited their potential applications. In this work, we develop a nucleation-controlled solution method to grow large size high-quality Cs3Bi2I9 perovskite single crystals (PSCs). Using the technique, we harvest some centimeter-sized single crystals and achieved high device performance. We find that X-ray detectors based on PSCs exhibit high sensitivity of 1652.3 μC Gyair−1 cm−2 and very low detectable dose rate of 130 nGyair s−1, both desired in medical diagnostics. In addition, its outstanding thermal stability inspires us to develop a high temperature X-ray detector with stable response at up to 100 °C. Furthermore, PSCs exhibit high X-ray imaging capability thanks to its negligible signal drifting and extremely high stability.

person 0.90 person 0.83 person 0.90 person 0.69 person 0.68 person 0.57 person 0.69 skis 0.59 person 0.53 person 0.53 a: RetinaNet (anchor-based, ResNeXt-101) b: Ours (anchor-based + FSAF, ResNet-50) Figure 1: Qualitative results of the anchor-based RetinaNet [22] using powerful ResNeXt-101 (left) and our detector with additional FSAF module using just ResNet-50 (right) under the same training and testing scale. Our FSAF module helps detecting hard objects like tiny person and flat skis with a less powerful backbone network. See Figure 7 for more examples. AbstractWe motivate and present feature selective anchor-free (FSAF) module, a simple and effective building block for single-shot object detectors. It can be plugged into singleshot detectors with feature pyramid structure. The FSAF module addresses two limitations brought up by the conventional anchor-based detection: 1) heuristic-guided feature selection; 2) overlap-based anchor sampling. The general concept of the FSAF module is online feature selection applied to the training of multi-level anchor-free branches. Specifically, an anchor-free branch is attached to each level of the feature pyramid, allowing box encoding and decoding in the anchor-free manner at an arbitrary level. During training, we dynamically assign each instance to the most suitable feature level. At the time of inference, the FSAF module can work jointly with anchor-based branches by outputting predictions in parallel. We instantiate this concept with simple implementations of anchor-free branches and online feature selection strategy. Experimental re-sults on the COCO detection track show that our FSAF module performs better than anchor-based counterparts while being faster. When working jointly with anchor-based branches, the FSAF module robustly improves the baseline RetinaNet by a large margin under various settings, while introducing nearly free inference overhead. And the resulting best model can achieve a state-of-the-art 44.6% mAP, outperforming all existing single-shot detectors on COCO.