Transcript Luis Guerra – Memristive Devices for Neuromorphic Computation

Memristive devices for
neuromorphic computation
Luís Guerra
IFIMUP-IN (Material Physics Institute of the University of
Porto – Nanoscience and Nanotechnology Institute)
New Challenges in the European Area: Young Scientist’s 1st
International Baku Forum
23rd of May, 2013
Outline
•
•
•
•
•
•
•
The Memristor
Applications
Neuromorphic Computation
Fabrication
Results
Willshaw Network
Conclusions
The Memristor
Theorized in 1971[1], physically achieved in 2008[2]:
- Two-terminal passive circuit element;
- Resistance depends on the history of
applied voltage or current;
- Self-crossing, pinched hysteretic I-V loop,
frequency dependent.
From [2]: D. B. Strukov, G. S. Snider, D. R.
Stewart, and R. S. Williams, Nature 453, 80
(2008).
𝜔1 ≫ 𝜔2 ≫ 𝜔3
From: Y. V. Pershin and M. Di Ventra, Advances
in Physics 60, 145–227 (2011)
[1] Chua, L. Memristor - The Missing Circuit Element. IEEE
Transactions On Circuit Theory CT-18, 507–519 (1971).
Applications
Resistive Random Access Memories (ReRAM)
- Non-volatile, reversible resistive switching;
- High-speed and high ON/OFF ratio;
- High-density;
- Possibly multi-level;
HP
Toshiba
Sandisk
Samsung
Panasonic
Neuromorphic computation – “the use of very-large-scale integration
(VLSI) systems, containing electronic analog circuits, to mimic neuro-biological
architectures present in the nervous system”
From: Mead, C. Neuromorphic electronic systems.
Proceedings of the IEEE 78, 1629–1636 (1990).
- Uncanny resemblance to biological synapses.
Neuromorphic Computation
Even the simplest brain is superior to a super computer,
the secret: ARCHITECTURE!
Human brain:
- 106 neurons / cm2
- 1010 synapses / cm2
- 2 mW / cm2
Total power consumption: 20 Watts
Memristors:
- Cheap
- Power efficient
- Small
From: Versace, M. & Chandler, B. The brain of a new machine. Spectrum, IEEE (2010).
Fabrication
Two-terminal resistance switches, typically a thin-film metalinsulator-metal (MIM) stack:
Metals:
Device area:
Ag, Al, Cu,
1 – 100 μm2
- Ion-beam for film deposition;
Pt, Ru, Ti.
- Optical litography for microfrabrication.
Insulator:
HfO2
150 μm2
From: Strukov, D. B. & Kohlstedt, H. Resistive switching phenomena in thin
films: Materials, devices, and applications. MRS Bulletin 37, 108–114 (2012).
Results
20
20
1
2
15
15
10
1
Current (mA)
Current (mA)
10
5
2
0
1
2
3
4
5
6
7
8
9
10
5
0
-5
-5
-10
Device area: 9 μm2
-10
-15
-20
-3
-2
-1
0
Voltage (V)
-
-15
1
2
3
-2
-1
0
Voltage (V)
Bipolar switching;
SET (HRS to LRS) and RESET (LRS to HRS) processes;
SET current compliance;
Loss of hysteresis with consecutive loops.
1
2
Results
20
10
0
0
-10
-10
-20
-20
-30
-40
Inset showing SETs in detail
1000
100
-30
10
1
-40
0.1
Current (mA)
1
2
3
4
5
6
7
Current (mA)
Current (mA)
10
-50
-60
0.01
1E-3
1E-4
1E-5
1E-6
-70
-50
1E-7
1E-8
Device area: 1
-60
μm2
1E-9
-80
0
2
4
6
8
10 12
14 16 18
20 22
Voltage (V)
-90
-70
-5
0
5
10
15
20
-5
Voltage (V)
- Bipolar switching;
- SET current compliance;
- High reset current / high Vset variability;
0
5
10
Voltage (V)
15
20
Willshaw Network
Associative memory mapping an input
vector into an output vector via a matrix
of binary synapses (memristors);
Nanodevices have high defect rates
Work around them!
Study of Stuck-at-0 (OFF) and Stuck-at-1 (ON) defects.
Capacity and robustness to noise can be improved by adjusting
the current readout threshold, according to the type of
predominant defect.
Conclusions
Memristor open possibilities for applications in:
- ReRAM and Neuromorphic computation, among others.
Key features of memristors:
- Resemblance to biological synapses;
- High scalability, below 10 nm;
- CMOS compatible;
- Fast, non-volatile, electrical switching;
- Low power consumption;
- Cheap.
Acknowledgments:
J. Ventura, C. Dias, P. Aguiar, J. Pereira, S. Freitas, P. P. Freitas
Thank you for your attention