The popularity and widespread usage of machine learning (ML) hardware have created challenges for its intellectual property (IP) protection. Logic locking is a widely used technique for IP protection but has received little attention in error-resilient applications such as ML hardware modules. This work investigates the effectiveness of logic locking when applied to tree-based ML circuits and reveals a critical vulnerability that undermines its effectiveness for single-label ML classifiers. We propose a logic locking scheme to eliminate the vulnerabilities in decision trees (DTs) and random forests (RFs) circuits. In our extensive simulation involving 16 DTs and 16 RFs, our solution consistently thwarts the vulnerability. We further evaluated the security of our approach by considering different obfuscation percentages and launching state-of-the-art oracle-less attacks on logic locking. Our method proves resilient, indicating that by fixing the identified vulnerability, we did not introduce new attack vectors. Further, our investigation indicates that DT/RF accelerators are significantly less vulnerable to oracle-less attacks compared to exact circuits. Overall, our work lays the foundation for future investigations into the effectiveness of logic locking for ML circuits.

This work explores the use of asynchronous approximate multiply-accumulate (MAC) operators and research ways to alleviate the inherent area overhead of such circuits, while leveraging on asynchronous circuits advantages. It analyzes three approximate MAC architectures with varying error rates and area trade-offs. Accurate and approximate, synchronous and asynchronous MAC operators are compared. Experiments show it is feasible to decrease the area overhead of the accurate asynchronous MAC from 8.1$\times$ down to 1.6$\times$ by recurring to approximate multipliers, for varying controlled error rates. The use of asynchronous quasi-delay insensitive (QDI) circuits allows applying extensive voltage scaling to all asynchronous MAC operators. Power and energy per operation can thus be significantly reduced, achieving savings of up to 2.66$\times$ in power and 3.17$\times$ in energy per operation when compared to the accurate asynchronous MAC.

Temperature rising is an unavoidable effect on VLSI and has always been a critical issue in any system-on-chip – especially when targeting compute-intensive applications. This effect increases the delay in hardware accelerators, resulting in timing errors due to unsustainable clock frequency, whose impact must be carefully evaluated on design time to measure the performance degradation of the hardware accelerator. Further, a hardware operating at a higher temperature accelerates device aging, which incurs in more timing errors. This issue is usually addressed with the inclusion of timing guardbands that compensate for the deleterious effects of temperature, ensuring the hardware accelerator works within a reliable zone, i.e., without any timing errors caused by temperature effects at runtime. However, guardbands directly result in considerable performance and efficiency losses because the circuit will be clocked at a frequency lower than its full potential. Accelerators on edge devices often dismiss such guardbands to explore the full potential of the designed circuits, posing an enormous design challenge as this approach requires a careful evaluation of the impact of timing errors on the quality of the target applications. Many algorithms, such as in multimedia and machine learning applications, are capable of tolerating hardware errors. Yet, these algorithms have a dynamic behavior (i.e., closed-loop) where a timing error can be propagated, affecting subsequent steps. Measuring the degradation-induced errors in these applications is very challenging given that an accurate gate-level simulation to investigate degradation-induced timing errors needs to be coupled dynamically with a system-level simulator to unveil how induced errors in the underlying hardware ultimately impact the algorithm execution in the hardware accelerator.This is the first work to achieve this goal. State-of-the-art works have studied accelerators under timing-errors when removing (or narrowing) guardbands. However, their approach was suitableonly for open-loop hardware accelerators which are entirely agnostic of complex interactions of the algorithms. Unlike prior work, this paper investigates temperature- and aging-induced timing-errors in the joint accelerator-algorithm interactions and their runtime impacts. Our framework investigates aging effects across the different layers starting from transistor physics all the way up to the algorithm layer. The hardware accelerator employed as a case study in this work is the sum of absolute differences (SAD), which is the most compute-intensive accelerator on commercial video encoder for mobile applications. Our results demonstrate the runtime behavior impacts of three advanced block-matching algorithms of the video encoder in a joint operation by a SAD accelerator under timing-errors induced by temperature and aging effects considering a 14nm FinFET technology.

Tree-based machine learning models, like random forests and decision trees, are low-complexity solutions that provide an efficient prediction for a wide range of applications. These models are particularly interesting for energy-constrained platforms since they can be implemented with simple logical operations. Tree-based accelerators are also intrinsically resilient to errors, and this can be leveraged to boost energy efficiency with approximate computing techniques. The key operations in these models are comparisons to constants, making comparators excellent candidates for approximation. This paper presents a technique to approximate comparisons to constants called C2PAx, which is capable of reducing the area and energy of tree-based accelerators. The method consists in finding alternative constants that reduce circuit area while keeping an efficient prediction performance. It is also shown that the selection of the constant parameters directly influences both hardware complexity and model performance, demanding cross-layer optimization. For that, we extend an existing framework that generates VLSI tree-based accelerators, inserting our approximation proposal that allows selecting the constant parameters that maximize energy efficiency at the cost of minor accuracy drops. Simulation results demonstrate that C2PAx outperforms the Don’t Care logic approximation technique when accuracy and energy are jointly considered. C2PAx trades accuracy for significant reductions in the VLSI area, power, delay, and energy consumption compared to precise models.

This paper presents a comprehensive design space exploration for boosting energy efficiency of a fast Fourier transform (FFT) VLSI accelerator, exploiting several approximate multipliers (AxM) combined with approximate adder (AxA) circuits. The FFT hardware herein presented consists of a fixed-point sequential architecture using a radix-2 butterfly with decimation in time. We explore a set of AxMs – namely Dynamic Range Unbiased (DRUM), Rounding-based Approximate (RoBA), leading one Bit-based Approximate (LoBA), and Truncated approach – jointly with the LOA, ETA-I, CopyA, CopyB, Trunc0, Trunc1 approximate adders. The approximate arithmetic operators are used in the butterfly kernel with exploration of the approximation levels (for the${L}$and${K}$least-significant bits, respectively, for the AxM and AxA), aiming at discovering the most energy-efficient configuration under a design-time QoR constraint. The mean square error and peak signal-to-noise ratio metrics define which approximate levels combining${L}$and${K}$variations will enable the FFT to process signals to generate spectrograms without significant losses. Our results show that the LoBA multiplier with$L$=8 together with the LOA, Trunc1 and Trunc0, at different approximation levels, provide most energy savings with controllable quality degradation, presenting a minimum decrease of 20.2% in power dissipation without degrading the spectrogram generation quality.

Logic synthesis is a fundamental step in hardware design whose goal is to find structural representations of Boolean functions while minimizing delay and area. If the function is completely-specified, the implementation accurately represents the function. If the function is incompletely-specified, the implementation has to be true only on the care set. While most of the algorithms in logic synthesis rely on SAT and Boolean methods to exactly implement the care set, we investigate learning in logic synthesis, attempting to trade exactness for generalization. This work is directly related to machine learning where the care set is the training set and the implementation is expected to generalize on a validation set. We present learning incompletely-specified functions based on the results of a competition conducted at IWLS 2020. The goal of the competition was to implement 100 functions given by a set of care minterms for training, while testing the implementation using a set of validation minterms sampled from the same function. We make this benchmark suite available and offer a detailed comparative analysis of the different approaches to learning.

This work presents an analysis of state-of-the-art learning-based image compression techniques. We compare 8 models available in the Tensorflow Compression package in terms of visual quality metrics and processing time, using the KODAK data set. The results are compared with the Better Portable Graphics (BPG) and the JPEG2000 codecs. Results show that JPEG2000 has the lowest execution times compared with the fastest learning-based model, with a speedup of $1.46 \times$ in compression and $30 \times$ in decompression. However, the learning-based models achieved improvements over JPEG2000 in terms of quality, specially for lower bitrates. Our findings also show that BPG is more efficient in terms of PSNR, but the learning models are better for other quality metrics, and sometimes even faster. The results indicate that learning-based techniques are promising solutions towards a future mainstream compression method.

This work evaluates the use of Decision Trees (DTs) methods for a fast logic minimization of Boolean functions. The proposed DT approach is compared to traditional Espresso logic minimizer and the minimization algorithms available in the ABC tool. The methods are compared with respect to the execution time, number of nodes and number of logic levels. The DT methods proved to be a faster alternative, reducing time by an average of 52% and 5.5% when compared to Espresso and ABC respectively, while keeping competitive results in terms of AIG depth and number of nodes. Additionally, in order to obtain smaller circuits at the cost of approximate results we tested DTs with limited tree depth. The trade-offs between synthesis time, circuit area and accuracy are also discussed. Compared to ABC, limiting the maximum tree depth leads to time savings of up to 52%, up to 86% less number of nodes, and up to 48% lower AIG depth, while maintaining acceptable accuracy results.

The use of Machine Learning techniques in battery-powered devices has increased in recent years. Therefore, evaluating power-accuracy trade-offs of inference models has become important. This work explores decision trees architectures, analyzing the effects of model complexity and approximation in power and accuracy. By quantizing the inputs to limited widths, we increased the accuracy of the models in up to 8.7% compared to precise cases. In terms of power, the variation in model complexity was more significant than in input width, as we obtained reductions of up to 97% by decreasing the tree depth, compared to 88% when decreasing the width.