Quantifying Synergistic Information

Citation

Griffith, Virgil (2014) Quantifying Synergistic Information. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/ZS2T-XQ55. https://resolver.caltech.edu/CaltechTHESIS:12132013-161604752

Abstract

Within the microcosm of information theory, I explore what it means for a system to be functionally irreducible. This is operationalized as quantifying the extent to which cooperative or "synergistic" effects enable random variables X₁, ... , X_n to predict (have mutual information about) a single target random variable Y . In Chapter 1, we introduce the problem with some emblematic examples. In Chapter 2, we show how six different measures from the existing literature fail to quantify this notion of synergistic mutual information. In Chapter 3 we take a step towards a measure of synergy which yields the first nontrivial lowerbound on synergistic mutual information. In Chapter 4, we find that synergy is but the weakest notion of a broader concept of irreducibility. In Chapter 5, we apply our results from Chapters 3 and 4 towards grounding Giulio Tononi’s ambitious φ measure which attempts to quantify the magnitude of consciousness experience.

Item Type:

Thesis (Dissertation (Ph.D.))

Subject Keywords:

information theory, synergy, consciousness, irreducibility

Degree Grantor:

California Institute of Technology

Division:

Engineering and Applied Science

Major Option:

Computation and Neural Systems

Awards:

Demetriades-Tsafka-Kokkalis Prize in Entrepreneurship or Related Fields, 2009

Thesis Availability:

Public (worldwide access)

Research Advisor(s):

Koch, Christof

Group:

Koch Laboratory (KLAB)

Thesis Committee:

Perona, Pietro (chair)
Koch, Christof
Ho, Tracey C.
Beck, James L.
Bruck, Jehoshua

Defense Date:

2 December 2013

Non-Caltech Author Email:

i (AT) virgil.gr

Additional Information:

Those interested in reading my thesis are encouraged to instead read the papers on arxiv.org. They cover the same material in more "bite-size chunks".

Funders:

Funding Agency	Grant Number
DOE CSGF (Department of Energy Computational Science Graduate Fellowship)	UNSPECIFIED
Department of Homeland Security Graduate Fellowship	UNSPECIFIED
Paul G. Allen Family Foundation	UNSPECIFIED

Record Number:

CaltechTHESIS:12132013-161604752

Persistent URL:

https://resolver.caltech.edu/CaltechTHESIS:12132013-161604752

DOI:

10.7907/ZS2T-XQ55

Related URLs:

URL	URL Type	Description
http://arxiv.org/abs/1205.4265	arXiv	Quantifying Synergistic Mutual Information - Adapted for ch. 2 and 3
http://arxiv.org/abs/1311.7442	arXiv	Irreducibility is Minimum Synergy Among Parts - Adapted for ch. 4
http://arxiv.org/abs/1310.1538	arXiv	Intersection Information based on Common Randomness - Adapted for ch. 3
http://arxiv.org/abs/1401.0978	arXiv	A Principled Infotheoretic φ-like Measure - Adapted for ch. 5