skip to main content

Title: Reversible Gating Architecture for Rare Failure Detection of Analog and Mixed-Signal Circuits
Due to the growing complexity and numerous manufacturing variation in safety-critical analog and mixed-signal (AMS) circuit design, rare failure detection in the high-dimensional variational space is one of the major challenges in AMS verification. Efficient AMS failure detection is very demanding with limited samples on account of high simulation and manufacturing cost. In this work, we combine a reversible network and a gating architecture to identify essential features from datasets and reduce feature dimension for fast failure detection. While reversible residual networks (RevNets) have been actively studied for its restoration ability from output to input without the loss of information, the gating network facilitates the RevNet to aim at effective dimension reduction. We incorporate the proposed reversible gating architecture into Bayesian optimization (BO) framework to reduce the dimensionality of BO embedding important features clarified by gating fusion weights so that the failure points can be efficiently located. Furthermore, we propose a conditional density estimation of important and non-important features to extract high-dimensional original input features from the low-dimension important features, improving the efficiency of the proposed methods. The improvements of our proposed approach on rare failure detection is demonstrated in AMS data under the high-dimensional process variations.  more » « less
Award ID(s):
Author(s) / Creator(s):
; ;
Date Published:
Journal Name:
IEEE/ACM Design Automation Conference
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. null (Ed.)
    Due to the extreme scarcity of customer failure data, it is challenging to reliably screen out those rare defects within a high-dimensional input feature space formed by the relevant parametric test measurements. In this paper, we study several unsupervised learning techniques based on six industrial test datasets, and propose to train a more robust unsupervised learning model by self-labeling the training data via a set of transformations. Using the labeled data we train a multi-class classifier through supervised training. The goodness of the multi-class classification decisions with respect to an unseen input data is used as a normality score to defect anomalies. Furthermore, we propose to use reversible information lossless transformations to retain the data information and boost the performance and robustness of the proposed self-labeling approach. 
    more » « less
  2. Set representation has become ubiquitous in deep learning for modeling the inductive bias of neural networks that are insensitive to the input order. DeepSets is the most widely used neural network architecture for set representation. It involves embedding each set element into a latent space with dimension L, followed by a sum pooling to obtain a whole-set embedding, and finally mapping the whole-set embedding to the output. In this work, we investigate the impact of the dimension L on the expressive power of DeepSets. Previous analyses either oversimplified high-dimensional features to be one-dimensional features or were limited to analytic activations, thereby diverging from practical use or resulting in L that grows exponentially with the set size N and feature dimension D. To investigate the minimal value of L that achieves sufficient expressive power, we present two set-element embedding layers: (a) linear + power activation (LP) and (b) linear + exponential activations (LE). We demonstrate that L being poly(N,D) is sufficient for set representation using both embedding layers. We also provide a lower bound of L for the LP embedding layer. Furthermore, we extend our results to permutation-equivariant set functions and the complex field. 
    more » « less
  3. null (Ed.)
    The invasive method of fetal electrocardiogram (fECG) monitoring is widely used with electrodes directly attached to the fetal scalp. There are potential risks such as infection and, thus, it is usually carried out during labor in rare cases. Recent advances in electronics and technologies have enabled fECG monitoring from the early stages of pregnancy through fECG extraction from the combined fetal/maternal ECG (f/mECG) signal recorded non-invasively in the abdominal area of the mother. However, cumbersome algorithms that require the reference maternal ECG as well as heavy feature crafting makes out-of-clinics fECG monitoring in daily life not yet feasible. To address these challenges, we proposed a pure end-to-end deep learning model to detect fetal QRS complexes (i.e., the main spikes observed on a fetal ECG waveform). Additionally, the model has the residual network (ResNet) architecture that adopts the novel 1-D octave convolution (OctConv) for learning multiple temporal frequency features, which in turn reduce memory and computational cost. Importantly, the model is capable of highlighting the contribution of regions that are more prominent for the detection. To evaluate our approach, data from the PhysioNet 2013 Challenge with labeled QRS complex annotations were used in the original form, and the data were then modified with Gaussian and motion noise, mimicking real-world scenarios. The model can achieve a F1 score of 91.1% while being able to save more than 50% computing cost with less than 2% performance degradation, demonstrating the effectiveness of our method. 
    more » « less
  4. Larger networks generally have greater representational power at the cost of increased computational complexity. Sparsifying such networks has been an active area of research but has been generally limited to static regularization or dynamic approaches using reinforcement learning. We explore a mixture of experts (MoE) approach to deep dynamic routing, which activates certain experts in the network on a per-example basis. Our novel DeepMoE architecture increases the representational power of standard convolutional networks by adaptively sparsifying and recalibrating channel-wise features in each convolutional layer. We employ a multi-headed sparse gating network to determine the selection and scaling of channels for each input, leveraging exponential combinations of experts within a single convolutional network. Our proposed architecture is evaluated on four benchmark datasets and tasks, and we show that Deep-MoEs are able to achieve higher accuracy with lower computation than standard convolutional networks. 
    more » « less
  5. Deep learning leverages multi-layer neural networks architecture and demonstrates superb power in many machine learning applications. The deep denoising autoencoder technique extracts better coherent features from the seismic data. The technique allows us to automatically extract low-dimensional features from high dimensional feature space in a non-linear, data-driven, and unsupervised way. A properly trained denoising autoencoder takes a partially corrupted input and recovers the original undistorted input. In this paper, a novel autoencoder built upon the deep residual network is proposed to perform noise attenuation on the seismic data. We evaluate the proposed method with synthetic datasets and the result confirms the effective denoising performance of the proposed approach. 
    more » « less