A Mathematical Formalization of Hierarchical Temporal Memory's Spatial Pooler

Mnatzaganian, James; Fokoué, Ernest; Kudithipudi, Dhireesha

doi:10.48550/arxiv.1601.06116

Cited by 2 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since we focused sequence learning in this paper, the proximal synapses were fixed during learning. In principle, the proximal synapses can also adapt continuously during learning according to a spatial competitive learning rule (Hawkins et al, 2011;Mnatzaganian et al, 2016).…”

Section: Htm Activation and Learning Rulesmentioning

confidence: 99%

Continuous Online Sequence Learning with an Unsupervised Neural Network Model

2016

View full text Add to dashboard Cite

The ability to recognize and predict temporal sequences of sensory inputs is vital for survival in natural environments. Based on many known properties of cortical neurons, hierarchical temporal memory (HTM) sequence memory is recently proposed as a theoretical framework for sequence learning in the cortex. In this paper, we analyze properties of HTM sequence memory and apply it to sequence learning and prediction problems with streaming data. We show the model is able to continuously learn a large number of variable-order temporal sequences using an unsupervised Hebbian-like learning rule. The sparse temporal codes formed by the model can robustly handle branching temporal sequences by maintaining multiple predictions until there is sufficient disambiguating evidence. We compare the HTM sequence memory with other sequence learning algorithms, including statistical methods: autoregressive integrated moving average (ARIMA), feedforward neural networks: online sequential extreme learning machine (ELM), and recurrent neural networks: long short-term memory (LSTM) and echo-state networks (ESN), on sequence prediction problems with both artificial and real-world data. The HTM model achieves comparable accuracy to other state-of-the-art algorithms. The model also exhibits properties that are critical for sequence learning, including continuous online learning, the ability to handle multiple predictions and branching sequences with high order statistics, robustness to sensor noise and fault tolerance, and good performance without task-specific hyperparameters tuning. Therefore the HTM sequence memory not only advances our understanding of how the brain may solve the sequence learning problem, but is also applicable to a wide range of real-world problems such as discrete and continuous sequence prediction, anomaly detection, and sequence classification.

show abstract

Section: Htm Activation and Learning Rulesmentioning

confidence: 99%

Continuous Online Sequence Learning with an Unsupervised Neural Network Model

2016

View full text Add to dashboard Cite

show abstract

“…• detraction of 1 on single synapse when multiple patterns activate the same column with opposite input value on single synapse (( • permanence boosting of inputs of columns activated by different patterns, harmful effect but if duty cycle big enough (almost bigger than number of patterns), inputs will be learned • attraction (equations above) -speeding up learning • detraction (equations above) -slowing down learning In both presented situations (single and multi pattern learning) constant inhibition radius is used. According to the original Numenta algorithm the fluctuations of inhibition radius should decrease during learning process [8], but there is possibility that inhibition radius never converge to constant value. In our approach the constant inhibition radius allows to show convergence of learning process.…”

Section: N Convergence Of Spmentioning

confidence: 99%

Formal Analysis of HTM Spatial Pooler Performance Under Predefined Operation Conditions

Pietroń¹,

Wielgosz²,

Wiatr³

2016

Lecture Notes in Computer Science

View full text Add to dashboard Cite

This paper introduces mathematical formalism for Spatial (SP) of Hierarchical Temporal Memory (HTM) with a spacial consideration for its hardware implementation. Performance of HTM network and its ability to learn and adjust to a problem at hand is governed by a large set of parameters. Most of parameters are codependent which makes creating efficient HTM-based solutions challenging. It requires profound knowledge of the settings and their impact on the performance of system. Consequently, this paper introduced a set of formulas which are to facilitate the design process by enhancing tedious trial-and-error method with a tool for choosing initial parameters which enable quick learning convergence. This is especially important in hardware implementations which are constrained by the limited resources of a platform.The authors focused especially on a formalism of Spatial Pooler and derive at the formulas for quality and convergence of the model. This may be considered as recipes for designing efficient HTM models for given input patterns.

show abstract

A Mathematical Formalization of Hierarchical Temporal Memory's Spatial Pooler

Cited by 2 publications

References 0 publications

Continuous Online Sequence Learning with an Unsupervised Neural Network Model

Continuous Online Sequence Learning with an Unsupervised Neural Network Model

Formal Analysis of HTM Spatial Pooler Performance Under Predefined Operation Conditions

Contact Info

Product

Resources

About