Upload
others
View
12
Download
0
Embed Size (px)
Citation preview
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
BENG 207 Special Topics in Bioengineering
Neuromorphic Integrated Bioelectronics
Week 8: Energy Conservation
Gert Cauwenberghs
Department of Bioengineering UC San Diego
http://isn.ucsd.edu/courses/beng207
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
BENG 207 Neuromorphic Integrated Bioelectronics
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
Resonant Adiabatic Energy Recovery
reso
nanc
e
(capacitive load)
Resonant adiabatic energy recovery in computing: Kerneltron support vector machine for visual pattern recognition with resonant hot clock adiabatic energy recovery in charge-domain processing-in-memory computing at 1 fJ of energy per multiply-accumulate [Karakiewicz et al, 2017, 2012]. Energy recovery logic (ERL) CMOS adiabatic line drivers recover 98% of the CV2 electrostatic energy in the charge-mode array.
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
Resonant Adiabatic Energy Recovery
Resonant adiabatic energy recovery in communication: Cyclic On-Off Keying (COOK) modulation for wireless power and telemetry offers record bandwidth efficiency, allowing to transmit one bit of data every carrier cycle while simultaneously receiving RF power over the same high-Q inductive link [Ha et al, 2016].
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
SVM Pattern Recognition
ASP A/D Sensory Features
Digital Analog
Large-Margin Kernel Regression
Class Identification Kerneltron: massively parallel support vector “machine” in silicon (ESSCIRC’2002)
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
Trainable Modular Vision Systems: The SVM Approach
– Strong mathematical foundations in Statistical Learning Theory (Vapnik, 1995)
– The training process selects a small fraction of prototype support vectors from the data set, located at the margin on both sides of the classification boundary (e.g., barely faces vs. barely non-faces) SVM classification for
pedestrian and face object detection
Papageorgiou, Oren, Osuna and Poggio, 1998
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
Trainable Modular Vision Systems: The SVM Approach
– The number of support vectors, in relation to the number of training samples and the vector dimension, determine the generalization performance
– Both training and run-time performance are severely limited by the computational complexity of evaluating kernel functions
ROC curve for various image representations and dimensions
Papageorgiou, Oren, Osuna and Poggio, 1998
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
Kernels and Support Vector Machines
x X
)(⋅Φ
)()(
xXxX
Φ=
Φ= ii
),( ⋅⋅K ),()()( xxxx ii K=Φ⋅Φ
)),((sign bKyySi
iii += ∑∈
xxα
))()((sign byySi
iii +Φ⋅Φ= ∑∈
xxα
)()( xxXX Φ⋅Φ=⋅ ii
Mercer’s Condition
)(sign byySi
iii +⋅= ∑∈
XXα
Mercer, 1909; Aizerman et al., 1964 Boser, Guyon and Vapnik, 1992
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
)exp( 2σ
xx ⋅∝ i
Kernel Machines
– Gaussian (Radial Basis Function Networks)
– Sigmoid (Two-Layer Perceptron)
– Polynomial (Splines etc.) ν)1(),( xxxx ⋅+= iiK
)exp(),( 2
2
2σ
xxxx −−= iiK
)tanh(),( xxxx ⋅+= ii LK only for certain L
k
k
x1x
2x
11yα
22yαsign y
)),((sign bKyySi
iii += ∑∈
xxα
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
Parallel SVM Architecture
– Kernel inner-products are implemented by parallel matrix-vector multiplication (MVM).
– Silicon area and power dissipation are proportional to number of support vectors, favoring sparse SVM solutions.
– Sparsity in SVM training also guarantees proper generalization performance (Vapnik, 1995).
x i
x SIGN
α i
MVM SUPPORT VECTORS
INPUT y
KE
RN
EL K
(x ,x i ) )),((sign bKyy
Siiii += ∑
∈
xxα
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
Kerneltron III: Adiabatic Support Vector “Machine”
• 1.2 TMACS / mW – adiabatic resonant clocking conserves
charge energy – energy efficiency on par with human
brain (1015 SynOP/S at 15W)
Karakiewicz, Genov, and Cauwenberghs, VLSI’2006; CICC’2007
x i
x SIGN
α i MVM
SUPPORT VECTORS
INPUT y
KE
RN
EL K
(x ,x i )
)),((sign bKyySi
iii += ∑∈
xxα
Classification results on MIT CBCL face detection data
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
CMOS Logic vs. Adiabatic Computing
Energy recovery logic (ERL) (Y. Moon, JSSC’96)
• ‘Hot clock’ recycles energy – LC tank resonant clock
• Reversible computation 2
. CVddEdiss =
CMOS logic
• Dynamic energy dissipation
Vdd
C
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
Resonant Charge Energy Recovery
resonance
capacitive load
CID array
reso
nanc
e
(capacitive load)
Karakiewicz, Genov, and Cauwenberghs, IEEE JSSC, 2007
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
Incremental and Decremental SVM Learning
– Support Vector Machine training requires solving a linearly constrained quadratic programming problem in a number of coefficients equal to the number of data points.
– An incremental version, training one data point at at time, is obtained by solving the QP problem in recursive fashion, without the need for QP steps or inverting a matrix.
• On-line learning is thus feasible, with no more than L2 state variables, where L is the number of margin (support) vectors.
• Training time scales approximately linearly with data size for large, low-dimensional data sets.
– Decremental learning (adiabatic reversal of incremental learning) allows to directly evaluate the exact leave-one-out generalization performance on the training data.
– When the incremental inverse jacobian is (near) ill-conditioned, a direct L1-norm minimization of the α coefficients yields an optimally sparse solution.
Cauwenberghs and Poggio, 2001
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
Trajectory of coefficients a as a function of time during incremental learning, for 100 data points in the non-separable case, and using a Gaussian kernel.
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
SVM Learning Revisited
iCy
Q
iii
i
ii
ijij
ji
SVcM
b
∀≤≤≡
−=
∑
∑∑∑
,0 and 0:subject to
:min 21
2,
αα
αααεw
Soft-Margin SVM Classification (Cortes and Vapnik, 1995)
∑+=i
ibzgC )(:min 2
21
,w
wε
)( byz iii +⋅= Xw
Primal Formulation
Dual Formulation
KKT Conditions
margin lo
ss
potential coefficient
),(
)('
jijiij
ij
jiji
ii
KyyQ
byQzzCg
xx=
+=
−=
∑ α
α
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
Sequential On-Line SVM Learning Chakrabartty, Genov and Cauwenberghs, 2003
ccccc
cc
QzzzCgα
α
+≈
−=0
)('
zc0 = zc
Qcczc0>1
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
Sequential On-Line SVM Learning Chakrabartty, Genov and Cauwenberghs, 2003
zc0
zc0<1
zc = 1
Qcc
ccccc
cc
QzzzCgα
α
+≈
−=0
)('
margin support vector
αc < C
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
Sequential On-Line SVM Learning Chakrabartty, Genov and Cauwenberghs, 2003
zc0
zc0<1
zc < C
Qcc
ccccc
cc
QzzzCgα
α
+≈
−=0
)('
error support vector
αc = C
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
Effects of Sequential On-Line Learning and Finite Resolution
• Matched Filter Response
• On-Line Sequential Training
• Kerneltron II
test train
-1
+1
• Batch Training
• Floating-
Point Resolution
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
SVM Sequence Estimation
ASP A/D Sensory Features
Digital Analog
MAP Forward Decoding
Sequence Identification
Sub-microwatt speaker verification and phoneme recognition (NIPS’2004)
GiniSVM
X[n] X[n-1] X[n+1]
j
i
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
X[1]
q[1]
X[2]
q[2]
X[N]
q[N]
Generative (MLE) HMM
Density models (such as mixtures of Gaussians) require vast amounts of training data to reliably estimate parameters.
Discriminative (MAP) FDKM
X[1]
q[1]
X[2]
q[2]
X[N]
q[N]
Transition-based speech recognition (H. Bourlard and N. Morgan, 1994)
MAP forward decoding
Transition probabilities generated by large margin probability regressor
1 2
P(1|1,x) P(2|2,x)
P(1|2,x)
P(2|1,x)
MLE vs. MAP Sequence Estimation
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
GiniSVM
Forward Decoding Kernel Machines (FDKM)
– Forward decoding of posterior probabilities αi
– Transition probabilities Pij generated by SVM conditioned on input data X
X[n] X[n-1]
∑ −=j
jiji nnPn ]1[][][ αα
])[(])[,|(][ nXfnXjiPnP ijij ∝=
X[n+1]
j
i
Chakrabartty and Cauwenberghs (NIPS’2002)
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
GiniSVM Sparse Probability Regression Chakrabartty and Cauwenberghs, JMLR 2007
aaaHy
HCQ
Giniiii
i
iCGini
ijij
ji
kGini
bi
)1(4)( with C,0 and 0:subject to
)(:min 21
2,
−=≤≤≡
−=
∑
∑∑∑
γαα
αα αεw
Gini Quadratic Entropy
Huber Loss Function
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
GiniSVM Probability Regression
Gini quadratic entropy in SVM training leads to sparse, large-margin regression of class probabilities (Chakrabartty et al. 2002)
x s
x NORMALIZATION
λ s i MVM MVM
SUPPORT VECTORS
INPUT f i (x)
P
KE
RN
EL K (x ,x s
) (i | x)
bKyfiPSs
siisi +=∝ ∑
∈
),()()|( xxxx λ
1)|(0
1)|(
≤≤
=∑x
x
iP
iPi
GiniSVM
iX
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
FDKM Architecture
x s
x NORMALIZATION
λ i1 s
14 24x24
30x24 30x24
1 2 24
α j[n-1] α i[n]
24
MVM MVM
SUPPORT VECTORS
INPUT f i1 (x)
24 FORWARD DECODING
P i1 P i2 4 K
ER
NE
L K(x ,x s )
24x24 GiniSVM
X[n] X[n-1] X[n+1]
j
i
bKf sSs
sijij +=∑
∈
),()( xxx λ
∑ −=j
jiji nnPn ]1[][][ αα
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
6 X 4 GiniSVMs
INPU
T ST
AG
E
30
14 24
PROGRAMMING REGISTERS
OUTPUT BUS
PRO
GR
AM
MIN
G R
EGIS
TER
S
GiniSVM/FDKM Processor Chakrabartty and Cauwenberghs (NIPS’2004)
Silicon support vector machine (SVM) and forward decoding kernel machine (FDKM)
x s
x NORMALIZATION
λ i1 s
14 24x24
30x24 30x24
1 2 24
α j[n-1] α i[n] 24
MVM MVM
SUPPORT VECTORS
INPUT f i1 (x)
24 FORWARD DECODING
P i1 P i2 4
KE
RN
EL K
(x,x s )
24x24
GiniSVM
X[n] X[n-1] X[n+1]
j
i
– Sub-Microwatt Power – Subthreshold translinear MOS circuits – Programmable with floating-gate non-
volatile analog storage
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
FDKM Dynamic Sequence Detection (80 nW)
q1q2 q3 q4 q5 q6
q8 q9 q10 q11 q12 q13
q7x1x2 x3 x4 x5 x6
x6x5 x4 x3 x2 x1
Chakrabartty and Cauwenberghs (NIPS’2004)
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
FDKM Training Formulation
– Large-margin training of state transition probabilities, using regularized cross-entropy on the posterior state probabilities:
– Forward Decoding Kernel Machines (FDKM) decompose an upper bound of the regularized cross-entropy (by expressing concavity of the logarithm in forward recursion on the previous state):
which then reduces to S independent regressions of conditional probabilities, one for each outgoing state:
Chakrabartty and Cauwenberghs, 2002
∑∑∑∑−
=
−
=
−
=
−
=
−=1
0
1
0
221
1
0
1
0||][log][
S
j
S
iij
N
n
S
iii wnnyCH α
∑∑ ∑−
=
−
=
−
=
−=1
0
221
1
0
1
0
||][log][][S
iij
N
n
S
iijijj wnPnynCH
∑−
=
≥1
0
S
jjHH
]1[][ −= nCnC jj α
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
Recursive MAP Training of FDKM
Epoch 2
n n-1 n-2
Epoch 1
n-1 n
1
2
Epoch K
n n-1 n-k
time
1
2
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
Phonetic Experiments (TIMIT)
0102030405060708090100
V S F N SV Sil
FDKMStatic
Features: cepstral coefficients for Vowels, Stops, Fricatives, Semi-Vowels, and Silence
Kernel Map Recognition Rate
Chakrabartty and Cauwenberghs, 2002
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
Programmable Analog Filter Bank (Deng et al, 2004 )
FDKM Chip
Con
ditio
nal
Prob
abili
ties
Speech
Log
Spec
tral
Fe
atur
es
On-Chip TIMIT Phone Recognition
– 6 phones /t/n/r/ow/ah/eh/ from TIMIT corpus – Thresholded Mel-cepstral features from log-compressed analog
filterbank
Chakrabartty and Cauwenberghs (NIPS’2004)
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
On-Chip Speaker Verification (840nW)
– 1 speaker and 10 imposters from YOHO dataset
– 92% recognition accuracy on 48 true and 432 imposter out-of-sample utterances
– 352 support vectors (47% FDKM chip capacity)
– 840 nW power at 25msec frame rate
Correct Speaker Imposter
Chakrabartty and Cauwenberghs (NIPS’2004)
Gert Cauwenberghs [email protected] BENG 207 Neuromorphic Integrated Bioelectronics
BENG 207 Neuromorphic Integrated Bioelectronics