1 / 18

The Role of Constraints in Hebbian Learning Miller and MacKay 1994, Neural Comput: 101-126

The Role of Constraints in Hebbian Learning Miller and MacKay 1994, Neural Comput: 101-126. Outline Constraints on Hebbian Plasticity: Importance Types of Constraints Dynamic Effects of Constraints Biological evidence for Constraints Function of Constraints in Heterogeneous Networks.

hedya
Download Presentation

The Role of Constraints in Hebbian Learning Miller and MacKay 1994, Neural Comput: 101-126

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. The Role of Constraints in Hebbian Learning Miller and MacKay 1994, Neural Comput: 101-126

  2. Outline • Constraints on Hebbian Plasticity: • Importance • Types of Constraints • Dynamic Effects of Constraints • Biological evidence for Constraints • Function of Constraints in Heterogeneous Networks

  3. Hebb’s Rule τw dw/dt = vu = C·w (Correlation based learning rule) where C = uu or the input correlation matrix • Problem with Hebb’s Rule: • weights grow without bounds => instability • loss of selectivity to different patterns of input

  4. Solution: Introduce Competition (constraint that limits total synaptic strength over cell) • Multiplicative: synapse decays at rate proportional to its current strength • Subtractive: synapse decays at a fixed rate where n = (1,1,….1)T in the synaptic basis

  5. Types of Constraints the two methods can enforce: • Type 1 – Conserve total synaptic strength • or Hence, Hyperplane Constraint Surface

  6. Type 2 – Conserve sum-squared synaptic strength Hence, M2: γ(w) = w · Cw/ w · w Hypersphere Constraint Surface

  7. Dynamic Effects of Multiplicative and Subtractive Constraints +ve correlation in Hebb’s rule (e0 is close in direction to constraint vector n) Cw α n Cw α w Correlations oscillate in sign (e0 || to constraint surface e0·n = 0 e0 treated as zero sum vector) e0 always  constrained surface

  8. Theorem 1: Under a multiplicatively enforced constraint, if principal eigen vector of C is an interior fixed point it is stable. Interior fixed points that are nonprincipal eigenvectors are unstable. (saturation unnecessary) Theorem 2: Under an S1 constraint, if C has atleast two eigenvectors with positive eigenvalues, then any interior fixed point is unstable. Theorem 3: If i and j are indices in the synaptic basis, and Cii > |Cij| then under S1 constraint, either all synapses (or all but one) are saturated in a stable final condition. S1 constraints lead to a zero sum vector that grows to complete saturation.

  9. Outcomes of development with and without Constraints RF has graded strengths RFs sharpened; Ocular dominance develops (final number of non-zero synapses α wtot/wmax)

  10. S1 constraints more appropriate to model Ocular Dominance than M1 constraints Let w1 and w2 be the synaptic weight vector from each input projection ws = w1 + w2 wd = w1 – w2 As inputs are symmetric, eigenvectors can be divided into sum eigenvectors: ws = eSa, wd = 0 (eigenval λSa) and wd = eDa, ws = 0 (eigenval λDa) Patterns of wd have zero total synaptic strength => wd grows freely under S1; wd suppressed under M1 (unless eDa is the principal eigenvector i.e. λDa > λSa – only possible with –ve correlations between inputs from two eyes!)

  11. Constraints applied to a full layer of output cells – Heuristic Approach

  12. Suggested Biological Implementation of Subtractive Constraints: • Limited capacity of metabolic supply to synapses (constant decay imposed on each synapse) • decay rate dependent on average degree of activation of cell However, S1 punishes weak synapses more than M1

  13. Evidence for another type of Constraint : Multiplicative Homeostatic Scaling in Cultured Networks Turrigiano and Nelson, Nature Rev Neuro. 5: 97-107 (2004)

  14. Physiology of Homeostatic Scaling ___ multiplicative scaling ….. random additive scaling ----- additive scaling Turrigiano et al, Nature 391: 892-896 (1998)

  15. Mechanisms of Homeostatic Scaling? Network level Single cell level

  16. Renart et al, Neuron 38: 473-485 (2004) Heterogeneous Cell Properties Homeostatic Scaling applied to Spatial Working Memory Homogeneous Cell Properties

  17. Homeostatic Scaling allows robust Spatial Memory Encoding in Heterogeneous Networks Isyn = gssgsyn(V-Vsyn)

  18. Summary • Constraints are required to clamp uncontrolled growth of Hebbian plasticity and to maintain input selectivity • Types of Constraints – S1, M1, M2 • Biological evidence for multiplicative scaling constraint • Multiplicative homeostatic scaling allows robust encoding in heterogeneous networks

More Related