More Like this

1. Monitoring the Health of Emerging Neural Network Accelerators with Cost-effective Concurrent Test

   Liu, Qi; Liu, Tao; Liu, Zihao; Wen, Wujie; Yang, Chengmo. (July 2020, IEEE/ACM Design Automation Conference (DAC))

   ReRAM-based neural network accelerator is a promising solution to handle the memory- and computation-intensive deep learning workloads. However, it suffers from unique device errors. These errors can accumulate to massive levels during the run time and cause significant accuracy drop. It is crucial to obtain its fault status in real-time before any proper repair mechanism can be applied. However, calibrating such statistical information is non-trivial because of the need of a large number of test patterns, long test time, and high test coverage considering that complex errors may appear in million-to-billion weight parameters. In this paper, we leverage the concept of corner data that can significantly confuse the decision making of neural network model, as well as the training algorithm, to generate only a small set of test patterns that is tuned to be sensitive to different levels of error accumulation and accuracy loss. Experimental results show that our method can quickly and correctly report the fault status of a running accelerator, outperforming existing solutions in both detection efficiency and cost. 

   more » « less
2. EdgeGuard: Robust and Fault-Aware Design for Resilient Edge Computing AI Accelerators

   https://doi.org/10.1145/3716368.3735153  

   Ahmed, Sabrina; Hoque, Khaza Anuarul; Carrion_Schafer, Benjamin (June 2025, ACM)

   Smaller transistor feature sizes have made integrated circuits (ICs) more vulnerable to permanent faults. This leads to short lifetimes and increased risk of faults that lead to catastrophic errors. Fortunately, Artificial Neural Networks (ANNs) are error resilient as their accuracies can be maintained through e.g., fault-aware re-training. One of the problems though with previous work is that they require a re-design in the individual neuron processing element structure in order to efficiently deal with these faults. In this work, we propose a novel architecture combined with a design flow that performs a fault-aware weight re-assignment in order to minimize the effect of permanent faults on the accuracy of ANNs mapped to AI accelerator without the need of time-consuming fault-aware re-training nor neuron processing elements re-design. In particular, we deal with Tensor Processing Units (TPUs) although our proposed approach is also extensible to any other architecture. Experimental results show that our proposed approach and can be efficiently executed on a fast dedicated hardware re-binding unit or on software. 

   more » « less
3. Exploiting Logic Locking for a Neural Trojan Attack on Machine Learning Accelerators

   https://doi.org/10.1145/3583781.3590242  

   Xu, Hongye; Liu, Dongfang; Merkel, Cory; Zuzak, Michael (June 2023, GLSVLSI '23: Proceedings of the Great Lakes Symposium on VLSI 2023)

   Logic locking has been proposed to safeguard intellectual property (IP) during chip fabrication. Logic locking techniques protect hardware IP by making a subset of combinational modules in a design dependent on a secret key that is withheld from untrusted parties. If an incorrect secret key is used, a set of deterministic errors is produced in locked modules, restricting unauthorized use. A common target for logic locking is neural accelerators, especially as machine-learning-as-a-service becomes more prevalent. In this work, we explore how logic locking can be used to compromise the security of a neural accelerator it protects. Specifically, we show how the deterministic errors caused by incorrect keys can be harnessed to produce neural-trojan-style backdoors. To do so, we first outline a motivational attack scenario where a carefully chosen incorrect key, which we call a trojan key, produces misclassifications for an attacker-specified input class in a locked accelerator. We then develop a theoretically-robust attack methodology to automatically identify trojan keys. To evaluate this attack, we launch it on several locked accelerators. In our largest benchmark accelerator, our attack identified a trojan key that caused a 74% decrease in classification accuracy for attacker-specified trigger inputs, while degrading accuracy by only 1.7% for other inputs on average. 

   more » « less
4. Noise Injection Adaption: End-to-End ReRAM Crossbar Non-ideal Effect Adaption for Neural Network Mapping

   https://doi.org/10.1145/3316781.3317870  

   He, Zhezhi; Lin, Jie; Ewetz, Rickard; Yuan, Jiann-Shiun; Fan, Deliang (June 2019, Proceedings of the 56th Annual Design Automation Conference 2019)

   In this work, we investigate various non-ideal effects (Stuck-At-Fault (SAF), IR-drop, thermal noise, shot noise, and random telegraph noise)of ReRAM crossbar when employing it as a dot-product engine for deep neural network (DNN) acceleration. In order to examine the impacts of those non-ideal effects, we first develop a comprehensive framework called PytorX based on main-stream DNN pytorch framework. PytorX could perform end-to-end training, mapping, and evaluation for crossbar-based neural network accelerator, considering all above discussed non-ideal effects of ReRAM crossbar together. Experiments based on PytorX show that directly mapping the trained large scale DNN into crossbar without considering these non-ideal effects could lead to a complete system malfunction (i.e., equal to random guess) when the neural network goes deeper and wider. In particular, to address SAF side effects, we propose a digital SAF error correction algorithm to compensate for crossbar output errors, which only needs one-time profiling to achieve almost no system accuracy degradation. Then, to overcome IR drop effects, we propose a Noise Injection Adaption (NIA) methodology by incorporating statistics of current shift caused by IR drop in each crossbar as stochastic noise to DNN training algorithm, which could efficiently regularize DNN model to make it intrinsically adaptive to non-ideal ReRAM crossbar. It is a one-time training method without the request of retraining for every specific crossbar. Optimizing system operating frequency could easily take care of rest non-ideal effects. Various experiments on different DNNs using image recognition application are conducted to show the efficacy of our proposed methodology. 

   more » « less
5. Automated Primary Hyperparathyroidism Screening with Neural Networks

   Ziems, Noah; Wu, Shaoen; Norman, Jim (January 2021, IEEE Globecom 2021)

   null (Ed.)

   Primary Hyperparathyroidism(PHPT) is a relatively common disease, affecting about one in every 1,000 adults. However, screening for PHPT can be difficult, meaning it often goes undiagnosed for long periods of time. While looking at specific blood test results independently can help indicate whether a patient has PHPT, often these blood result levels can all be within their respective normal ranges despite the patient having PHPT. Based on the clinic data from the real world, in this work, we propose a novel approach to screening PHPT with neural network (NN) architecture, achieving over 97% accuracy with common blood values as inputs. Further, we propose a second model achieving over 99% accuracy with additional lab test values as inputs. Moreover, compared to traditional PHPT screening methods, our NN models can reduce the false negatives of traditional screening methods by 99%. 

   more » « less