A theory of sequence memory in the neocortex · A theory of sequence memory in the neocortex Jeff...
Transcript of A theory of sequence memory in the neocortex · A theory of sequence memory in the neocortex Jeff...
A theory of sequence memory in the neocortexJeff Hawkins, Subutai Ahmad, Yuwei Cui and Chetan Surpur
[email protected], [email protected], [email protected]
We thank Scott Purdy, Alex Lavin, Matthew Larkum, Ahmed El Hady for helpful discussions.
Continuous learning from sequence streams
4. Works well on real-world problems
Poirazi P, Brannon T, Mel BW (2003) Neuron 37:989–999.Chklovskii, D. B., Mel, B. W., and Svoboda, K. (2004). Nature 431, 782–8. Hawkins, J., and Ahmad, S. (2015). arXiv:1511.00083 [q-bio.NC]Holtmaat A, Svoboda K (2009) Nat Rev Neurosci 10:647–658.Kanerva, P, Sparse Distributed Memory. The MIT Press, 1988Lamme, V. A., Supèr, H., and Spekreijse, H. (1998). Curr. Opin. Neurobiol. 8, 529–35.Losonczy, A., Makara, J. K., and Magee, J. C. (2008). Nature 452, 436–41. Major, G., Larkum, M. E., and Schiller, J. (2013). Annu. Rev. Neurosci. 36, 1–24. Piuri, V., J Parallel Distr Com., vol. 61, pp. 18-48.Smith SL, Smith IT, Branco T, Häusser M (2013) Nature 503:115–120.Vinje, W. E., and Gallant, J. L. (2002). J. Neurosci. 22, 2904–2915.
Streams of sensory inputs
...
1. Sequence learning is ubiquitous in cortex
References
Active Dendrites
Small sets of synapses in close proximity act as independent pattern detectors.
Detection of a pattern causes an NMDA spike and depolarization at the soma.
Depolarization acts as prediction, causing cell to fire earlier.
Learning occurs by growing new synapses via Hebbian learning rule.
Our code is open sourceWe believe in open research and full transparency. Numenta's research and algorithm code is part of the open-source project Numenta Platform for Intelligent Computing (NuPIC). As a fast growing project, NuPIC currently has more than 4,000 followers and more than 1000 forks on Github.
What is neural mechanism for sequence learning?
HTM Sequence Memory:1. Neurons learn to recognize hundreds of patterns using active dendrites.
2. Recognition of patterns act as predictions by depolarizing the cell without generating an immediate action potential.
3. A network of neurons with active dendrites forms a power-ful sequence memory
Acknowledgements
Sequence recognitionSequence predictionBehavior generation
X
A B
B
C
C
D
Y
Before learning
X B’’ C’’
D’
Y’’
After learning
A B’ C’
Same columns,but only one cell active per column after learning,
Active cells
Depolarized (predictive) cells
Inactive cells
Time
3. HTM network model for sequence learning
A sparse set of columns becomes activedue to intercolumn inhibition
High-order sequence Noise High-order sequence Noise ......
Task: sequence prediction with streams of high-order sequences
Check prediction at the end of each sequence
Learning complex high-order sequences
ABCD vs XBCY
0 5000 10000 15000 20000 25000Number of elements seen
0.0
0.2
0.4
0.6
0.8
1.0
Pre
dict
ion
Acc
urac
y
HTMLSTM window=1000LSTM window=3000LSTM window=9000
Example sequences:
G, I, H, E, C, D, AB, I, H, E, C, D, F
A, J, H, I, F, D, E, BC, J, H, I, F, D, E, G......
Modified sequences:
G, I, H, E, C, D, FB, I, H, E, C, D, A
A, J, H, I, F, D, E, GC, J, H, I, F, D, E, B......
HTM learns continuously, no batch training required.HTM is more robust and recovers more quickly.
High fault tolerance to neuron death
0.0 0.1 0.2 0.3 0.4 0.5 0.6Fraction of cell death
0.2
0.4
0.6
0.8
1.0
Acc
urac
y af
ter c
ell d
eath
HTMLSTM HTM is fault tolerant due to properties of
sparse distributed representations (Kaner-va 1988) and nonlinear dendritic properties of HTM neurons (Hawkins & Ahmad 2015).
Mea
n ab
solu
te p
erce
nt e
rror
Neg
ativ
e Lo
g-lik
elih
ood
Shift
ARIMAESN
LSTM10
00
LSTM30
00
LSTM60
00HTM
0.00
0.10
0.20
0.30
LSTM10
00
LSTM30
00
LSTM60
00HTM
0.0
0.5
1.0
1.5
2015-04-20 Monday
2015-04-21Tuesday
2015-04-22Wednesday
2015-04-23 Thursday
2015-04-24 Friday
2015-04-25 Saturday
2015-04-26 Sunday
0k
5k
10k
15k
20k
25k
Pas
seng
er C
ount
Data sources: http://www.nyc.gov/html/tlc/html/about/trip_record_data.shtml
Task: predict taxi passenger count in NYC
Apr 01 Apr 08 Apr 15 Apr 22 Apr 29 May 06
LSTM3000LSTM6000HTM
Mea
n ab
solu
te p
erce
nt e
rror
Neg
ativ
e Lo
g-lik
elih
ood
0.10
0.15
0.20
0.25
0.30
0.35
Mea
n ab
solu
te p
erce
nt e
rror
LSTM10
00
LSTM30
00
LSTM60
00HTM
0.00
0.02
0.04
0.06
0.08
0.10
0.12
LSTM10
00
LSTM30
00
LSTM60
00HTM
0.0
0.5
1.0
1.5
2.0
HTM has comparable performance to state-of-the-art algorithms
20% increase of weekday night traffic (9pm-11pm)
20% decrease of weekday morning traffic (7am-11am)
HTM quickly adapts to changes due to its ability to learn continuously
Major, Larkum and Schiller 2013
Activation rules: Select the top 2% of columns with strongest inputs on proximal dendrite as active columns
If any cell in an active column is predicted, only the predicted cells fire
If no cell in an active column is predicted, all cells in the column fire
Unsupervised Hebbian-like learning rules: If a depolarized cell becomes active subsequently, its active dendritic segment will be reinforced
If a depolarized cell does not become active, we apply a small decay to active segments of that cell
If no cell in an active column is predicted, the cell with the most activated segment gets reinforced
6. Testable predictions
Apical inputs predict entire sequences
Feedback biases network for sequence B’ C’ D’
Apical dendrites
Input CRepresentation C’
Input YDoes not match expectation
It has been speculated that feedback connections implement expectation or bias (Lamme et al., 1998). Our neuron model suggests a mechanism for top-down expectation in the cortex.
Feedback to the apical dendrites can predict multiple elements simultaneously.New feedforward input will be intepreted as part of the predicted sequence.
1) Sparser activations during a predictable sensory stream. (Vinje & Gallant 2002)2) Unanticipated inputs lead to a burst of activity correlated vertically within mini-columns.3) Neighboring mini-columns will not be correlated.4) Predicted cells need fast inhibition to inhibit nearby cells within mini-column.5) For predictable stimuli, dendritic NMDA spikes will be much more frequent than somatic action potentials. (Smith et al., 2013)6) Strong LTP in distal dendrites requires bAP and NMDA spike (Losonczy et al., 2008)7) Weak LTP (in the absence of NMDA spikes) in dendritic segments if a cluster of synapses become active followed by a bAP.8) Localized weak LTD when an NMDA spike is not followed by a bAP.
· Unsupervised learning· Quickly adapts to changes in data· Learns high-order structure in sequences· Robust and fault tolerant· Makes multiple simultaneous predictions· Works well on real-world problems· Accurate biological model
2. HTM neuron models active dendrites
Learning and activation rules
Before learning, Unexpected input After learning, Predicted input
Two separate sparse representations
Only one cell active per column,due to intracolumn inhibition
HTM exhibits many desirable features for sequence learning:
Active Dendrites
Pattern detectors Each cell can recognize 100’s of unique patterns
Feedforward
100’s of synapses “Classic” receptive field
Context
1,000’s of synapses Depolarize neuron “Predicted state”
Pyramidal neuron HTM neuron
Each segment on distal dendrite can recognize a particular pattern and activate dendrite(Poirazi et al., 2003; Palmer et al., 2014)
Proximal dendrite can recognize a particular pattern and activate cell
Switch to modified dataset
0.0 0.3 1.0 Synapse permanence
0 1 Synapse weight
Dendrite
Axon
Learning with synaptogenesis
Chklovskii, Mel, and Svoboda 2004Holtmaat and Svoboda 2009
5. Summary
Further implementation detail can be found at https://github.com/numenta/nupic
Injected change to the data
In contrast, LSTM and most other artificial neural networks are sensitive to loss of neurons or synapses (Piuri 2001)
February 2016