Training Fixed-Point Classifiers for On-Chip Low-Power Implementation

In this article, we develop several novel algorithms to train classifiers that can be implemented on chip with low-power fixed-point arithmetic with extremely small word length. These algorithms are based on Linear Discriminant Analysis (LDA), Support Vector Machine (SVM), and Logistic Regression (LR), and are referred to as LDA-FP, SVM-FP, and LR-FP, respectively. They incorporate the nonidealities (i.e., rounding and overflow) associated with fixed-point arithmetic into the offline training process so that the resulting classifiers are robust to these nonidealities. Mathematically, LDA-FP, SVM-FP, and LR-FP are formulated as mixed integer programming problems that can be robustly solved by the branch-and-bound methods described in this article. Our numerical experiments demonstrate that LDA-FP, SVM-FP, and LR-FP substantially outperform the conventional approaches for the emerging biomedical applications of brain decoding.

[1]  Jeffrey M. Hausdorff,et al.  Physionet: Components of a New Research Resource for Complex Physiologic Signals". Circu-lation Vol , 2000 .

[2]  Xin Li,et al.  Low-power hardware implementation of movement decoding for brain computer interface with reduced-resolution discrete cosine transform , 2014, 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[3]  David V. Anderson,et al.  Fixed-Point Signal Processing , 2009, Synthesis Lectures on Signal Processing.

[4]  Nicola Nicolici,et al.  Automated Range and Precision Bit-Width Allocation for Iterative Computations , 2011, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[5]  Eric Laciar,et al.  Automatic detection of drowsiness in EEG records based on multimodal analysis. , 2014, Medical engineering & physics.

[6]  Stephen P. Boyd,et al.  Tractable approximate robust geometric programming , 2007, Optimization and Engineering.

[7]  G. Nemhauser,et al.  Integer Programming , 2020 .

[8]  Anantha Chandrakasan,et al.  An 8-Channel Scalable EEG Acquisition SoC With Patient-Specific Seizure Classification and Recording Processor , 2013, IEEE Journal of Solid-State Circuits.

[9]  Anantha Chandrakasan,et al.  An 8-channel scalable EEG acquisition SoC with fully integrated patient-specific seizure classification and recording processor , 2012, 2012 IEEE International Solid-State Circuits Conference.

[10]  Miguel A. L. Nicolelis,et al.  Actions from thoughts , 2001, Nature.

[11]  Hai Zhou,et al.  Low-Power Optimization by Smart Bit-Width Allocation in a SystemC-Based ASIC Design Environment , 2007, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[12]  Mervyn V. M. Yeo,et al.  Can SVM be used for automatic EEG detection of drowsiness during car driving , 2009 .

[13]  Stephen P. Boyd,et al.  Convex Optimization , 2004, Algorithms and Theory of Computation Handbook.

[14]  David A. Clifton,et al.  Gaussian process regression in vital-sign early warning systems , 2012, 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[15]  Refet Firat Yazicioglu,et al.  A Configurable and Low-Power Mixed Signal SoC for Portable ECG Monitoring Applications , 2011, IEEE Transactions on Biomedical Circuits and Systems.

[16]  Wayne Luk,et al.  Wordlength optimization for linear digital signal processing , 2003, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..

[17]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[18]  M. Kringelbach,et al.  Translational principles of deep brain stimulation , 2007, Nature Reviews Neuroscience.

[19]  Robin C. Ashmore,et al.  An Electrocorticographic Brain Interface in an Individual with Tetraplegia , 2013, PloS one.

[20]  G. Moody,et al.  Development of the polysomnographic database on CD‐ROM , 1999, Psychiatry and clinical neurosciences.

[21]  Naveen Verma,et al.  A compressed-domain processor for seizure detection to simultaneously reduce computation and communication energy , 2012, Proceedings of the IEEE 2012 Custom Integrated Circuits Conference.

[22]  Hoi-Jun Yoo,et al.  A 259.6μW nonlinear HRV-EEG chaos processor with body channel communication interface for mental health monitoring , 2012, 2012 IEEE International Solid-State Circuits Conference.

[23]  Kenneth Sundaraj,et al.  Detecting Driver Drowsiness Based on Sensors: A Review , 2012, Sensors.

[24]  Charles Sodini,et al.  A Wearable Cardiac Monitor for Long-Term Data Acquisition and Analysis , 2013, IEEE Transactions on Biomedical Engineering.

[25]  Gene H. Golub,et al.  Matrix computations , 1983 .