At the dawn of the fourth industrial revolution, we are witnessing a fast and widespread adoption of artificial intelligence (AI) in our daily life, which contributes to accelerating the shift towards a more algorithmic society. However, even with such unprecedented advancements, a key impediment to the use of AI-based systems is that they often lack transparency. Indeed, the black-box nature of these systems allows powerful predictions, but it cannot be directly explained. This issue has triggered a new debate on explainable AI (XAI). A research field holds substantial promise for improving trust and transparency of AI-based systems. It is recognized as the sine qua non for AI to continue making steady progress without disruption. This survey provides an entry point for interested researchers and practitioners to learn key aspects of the young and rapidly growing body of research related to XAI. Through the lens of the literature, we review the existing approaches regarding the topic, discuss trends surrounding its sphere, and present major research trajectories.

Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI)

Eyeriss is an accelerator for state-of-the-art deep convolutional neural networks (CNNs). It optimizes for the energy efficiency of the entire system, including the accelerator chip and off-chip DRAM, for various CNN shapes by reconfiguring the architecture. CNNs are widely used in modern AI systems but also bring challenges on throughput and energy efficiency to the underlying hardware. This is because its computation requires a large amount of data, creating significant data movement from on-chip and off-chip that is more energy-consuming than computation. Minimizing data movement energy cost for any CNN shape, therefore, is the key to high throughput and energy efficiency. Eyeriss achieves these goals by using a proposed processing dataflow, called row stationary (RS), on a spatial architecture with 168 processing elements. RS dataflow reconfigures the computation mapping of a given shape, which optimizes energy efficiency by maximally reusing data locally to reduce expensive data movement, such as DRAM accesses. Compression and data gating are also applied to further improve energy efficiency. Eyeriss processes the convolutional layers at 35 frames/s and 0.0029 DRAM access/multiply and accumulation (MAC) for AlexNet at 278 mW (batch size $N = 4$ ), and 0.7 frames/s and 0.0035 DRAM access/MAC for VGG-16 at 236 mW ( $N = 3$ ).

Eyeriss: An Energy-Efficient Reconfigurable Accelerator for Deep Convolutional Neural Networks

Confidence calibration -- the problem of predicting probability estimates representative of the true correctness likelihood -- is important for classification models in many applications. We discover that modern neural networks, unlike those from a decade ago, are poorly calibrated. Through extensive experiments, we observe that depth, width, weight decay, and Batch Normalization are important factors influencing calibration. We evaluate the performance of various post-processing calibration methods on state-of-the-art architectures with image and document classification datasets. Our analysis and experiments not only offer insights into neural network learning, but also provide a simple and straightforward recipe for practical settings: on most datasets, temperature scaling -- a single-parameter variant of Platt Scaling -- is surprisingly effective at calibrating predictions.

On Calibration of Modern Neural Networks

/pdf/on-calibration-of-modern-neural-networks-ejfsky7uty.pdf

On calibration of modern neural networks

Neural networks are known to be vulnerable to adversarial examples: inputs that are close to natural inputs but classified incorrectly. In order to better understand the space of adversarial examples, we survey ten recent proposals that are designed for detection and compare their efficacy. We show that all can be defeated by constructing new loss functions. We conclude that adversarial examples are significantly harder to detect than previously appreciated, and the properties believed to be intrinsic to adversarial examples are in fact not. Finally, we propose several simple guidelines for evaluating future proposed defenses.

https://dl.acm.org/doi/pdf/10.1145/3128572.3140444

Adversarial Examples Are Not Easily Detected: Bypassing Ten Detection Methods

We trained a convolutional neural network (CNN) to map raw pixels from a single front-facing camera directly to steering commands. This end-to-end approach proved surprisingly powerful. With minimum training data from humans the system learns to drive in traffic on local roads with or without lane markings and on highways. It also operates in areas with unclear visual guidance such as in parking lots and on unpaved roads. 
The system automatically learns internal representations of the necessary processing steps such as detecting useful road features with only the human steering angle as the training signal. We never explicitly trained it to detect, for example, the outline of roads. 
Compared to explicit decomposition of the problem, such as lane marking detection, path planning, and control, our end-to-end system optimizes all processing steps simultaneously. We argue that this will eventually lead to better performance and smaller systems. Better performance will result because the internal components self-optimize to maximize overall system performance, instead of optimizing human-selected intermediate criteria, e.g., lane detection. Such criteria understandably are selected for ease of human interpretation which doesn't automatically guarantee maximum system performance. Smaller networks are possible because the system learns to solve the problem with the minimal number of processing steps. 
We used an NVIDIA DevBox and Torch 7 for training and an NVIDIA DRIVE(TM) PX self-driving car computer also running Torch 7 for determining where to drive. The system operates at 30 frames per second (FPS).

/pdf/end-to-end-learning-for-self-driving-cars-ru8bmvuyqf.pdf

End to End Learning for Self-Driving Cars

/pdf/multigenerational-birth-cohort-study-in-china-importance-1ndzfrkq.pdf

Multigenerational birth cohort study in China: importance, necessity and beyond

/pdf/correction-to-multigenerational-birth-cohort-study-in-china-2qtfwbhb.pdf

Correction to: Multigenerational birth cohort study in China: importance, necessity and beyond

An overview of methods of left and right foot motor imagery based on Tikhonov regularisation common spatial pattern

Supercritical fluid extraction (SCFE) is gaining significant interest as a green technology for the recycling of end-of-life waste electrical and electronic equipment (WEEE). Neodymium iron boron (NdFeB) magnets, which contain large quantities of critical rare-earth elements such as neodymium, praseodymium, and dysprosium, are widely used in wind turbines and electric/hybrid vehicles. Hence, they are considered a promising secondary resource for these elements when they reach their end-of-life. Previously, the SCFE process was developed for recycling WEEE, including NdFeB; however, the process mechanism remains unexplored. Here, density functional theory, followed by extended X-ray absorption fine structure and X-ray absorption near-edge structure analyses, are utilized to determine the structural coordination and interatomic interactions of complexes formed during the SCFE of the NdFeB magnet. The results indicate that Fe(II), Fe(III), and Nd(III) form Fe(NO3)2(TBP)2, Fe(NO3)3(TBP)2, and Nd(NO3)3(TBP)3 complexes, respectively. This theory-guided investigation elucidates the complexation chemistry and mechanism during the SCFE process by rigorously determining the structural models.

Investigating Metal-Tributyl Phosphate Complexes during Supercritical Fluid Extraction of the NdFeB Magnet Using Density Functional Theory and X-ray Absorption Spectroscopy.

Jiakai Zhang

Papers

End to End Learning for Self-Driving Cars

Multigenerational birth cohort study in China: importance, necessity and beyond

Correction to: Multigenerational birth cohort study in China: importance, necessity and beyond

An overview of methods of left and right foot motor imagery based on Tikhonov regularisation common spatial pattern

Investigating Metal-Tributyl Phosphate Complexes during Supercritical Fluid Extraction of the NdFeB Magnet Using Density Functional Theory and X-ray Absorption Spectroscopy.