skip to main content

This content will become publicly available on May 22, 2024

Title: Error Resilient Transformer Networks: A Novel Sensitivity Guided Approach to Error Checking and Suppression
Transformer networks have achieved remarkable success in Natural Language Processing (NLP) and Computer Vision applications. However, the underlying large volumes of Transformer computations demand high reliability and resilience to soft errors in processor hardware. The objective of this research is to develop efficient techniques for design of error resilient Transformer architectures. To enable this, we first perform a soft error vulnerability analysis of every fully connected layers in Transformer computations. Based on this study, error detection and suppression modules are selectively introduced into datapaths to restore Transformer performance under anticipated error rate conditions. Memory access errors and neuron output errors are detected using checksums of linear Transformer computations. Correction consists of determining output neurons with out-of-range values and suppressing the same to zero. For a Transformer with nominal BLEU score of 52.7, such vulnerability guided selective error suppression can recover language translation performance from a BLEU score of 0 to 50.774 with as much as 0.001 probability of activation error, incurring negligible memory and computation overheads.  more » « less
Award ID(s):
Author(s) / Creator(s):
; ;
Date Published:
Journal Name:
European Test Symposium
Page Range / eLocation ID:
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. The reliability of emerging neuromorphic compute fabrics is of great concern due to their widespread use in critical data-intensive applications. Ensuring such reliability is difficult due to the intensity of underlying computations (billions of parameters), errors induced by low power operation and the complex relationship between errors in computations and their effect on network performance accuracy. We study the problem of designing error-resilient neuromorphic systems where errors can stem from: (a) soft errors in computation of matrix-vector multiplications and neuron activations, (b) malicious trojan and adversarial security attacks and (c) effects of manufacturing process variations on analog crossbar arrays that can affect DNN accuracy. The core principle of error detection relies on embedded predictive neuron checks using invariants derived from the statistics of nominal neuron activation patterns of hidden layers of a neural network. Algorithmic encodings of hidden neuron function are also used to derive invariants for checking. A key contribution is designing checks that are robust to the inherent nonlinearity of neuron computations with minimal impact on error detection coverage. Once errors are detected, they are corrected using probabilistic methods due to the difficulties involved in exact error diagnosis in such complex systems. The technique is scalable across soft errors as well as a range of security attacks. The effects of manufacturing process variations are handled through the use of compact tests from which DNN performance can be assessed using learning techniques. Experimental results on a variety of neuromorphic test systems: DNNs, spiking networks and hyperdimensional computing are presented. 
    more » « less
  2. Manually writing parallel programs is difficult and error-prone. Automatic parallelization could address this issue, but profitability can be limited by not having facts known only to the programmer. A parallelizing compiler that collaborates with the programmer can increase the coverage and performance of parallelization while reducing the errors and overhead associated with manual parallelization. Unlike collaboration involving analysis tools that report program properties or make parallelization suggestions to the programmer, decompiler-based collaboration could leverage the strength of existing parallelizing compilers to provide programmers with a natural compiler-parallelized starting point for further parallelization or refinement. Despite this potential, existing decompilers fail to do this because they do not generate portable parallel source code compatible with any compiler of the source language. This paper presents SPLENDID, an LLVM-IR to C/OpenMP decompiler that enables collaborative parallelization by producing standard parallel OpenMP code. Using published manual parallelization of the PolyBench benchmark suite as a reference, SPLENDID's collaborative approach produces programs twice as fast as either Polly-based automatic parallelization or manual parallelization alone. SPLENDID's portable parallel code is also more natural than that from existing decompilers, obtaining a 39x higher average BLEU score. 
    more » « less
  3. Dual learning has attracted much attention in machine learning, computer vision and natural language processing communities. The core idea of dual learning is to leverage the duality between the primal task (mapping from domain X to domain Y) and dual task (mapping from domain Y to X) to boost the performances of both tasks. Existing dual learning framework forms a system with two agents (one primal model and one dual model) to utilize such duality. In this paper, we extend this framework by introducing multiple primal and dual models, and propose the multi-agent dual learning framework. Experiments on neural machine translation and image translation tasks demonstrate the effectiveness of the new framework. In particular, we set a new record on IWSLT 2014 German-to-English translation with a 35.44 BLEU score, achieve a 31.03 BLEU score on WMT 2014 English-to-German translation with over 2.6 BLEU improvement over the strong Transformer baseline, and set a new record of 49.61 BLEU score on the recent WMT 2018 English-to-German translation. 
    more » « less
  4. Online reinforcement learning (RL) based systems are being increasingly deployed in a variety of safety-critical applications ranging from drone control to medical robotics. These systems typically use RL onboard rather than relying on remote operation from high-performance datacenters. Due to the dynamic nature of the environments they work in, onboard RL hardware is vulnerable to soft errors from radiation, thermal effects and electrical noise that corrupt the results of computations. Existing approaches to on-line error resilience in machine learning systems have relied on availability of the large training datasets to configure resilience parameters, which is not necessarily feasible for online RL systems. Similarly, other approaches involving specialized hardware or modifications to training algorithms are difficult to implement for onboard RL applications. In contrast, we present a novel error resilience approach for online RL that makes use of running statistics collected across the (real-time) RL training process to configure error detection thresholds without the need to access a reference training dataset. In this methodology, statistical concentration bounds leveraging running statistics are used to diagnose neuron outputs as erroneous. These erroneous neurons are then set to zero (suppressed). Our approach is compared against the state of the art and validated on several RL algorithms involving the use of multiple concentration bounds on CPU as well as GPU hardware. 
    more » « less
  5. Approximate computing (AC) leverages the inherent error resilience and is used in many big-data applications from various domains such as multimedia, computer vision, signal processing, and machine learning to improve systems performance and power consumption. Like many other approximate circuits and algorithms, the memory subsystem can also be used to enhance performance and save power significantly. This paper proposes an efficient and effective systematic methodology to construct an approximate non-volatile magneto-resistive RAM (MRAM) framework using consumer-off-the-shelf (COTS) MRAM chips. In the proposed scheme, an extensive experimental characterization of memory errors is performed by manipulating the write latency of MRAM chips which exploits the inherent (intrinsic/extrinsic process variation) stochastic switching behavior of magnetic tunnel junctions (MTJs). The experimental results, involving error-resilient image compression and machine learning applications, reveal that the proposed AC framework provides a significant performance improvement and demonstrates a reduction in MRAM write energy of ~47.5% on average with negligible or no loss in output quality. 
    more » « less