Almaden Institute

    April 11-12, 2007:
    Navigating Complexity: Doing more with less

Abstract: This talk focuses on architectural and organizational principles of networked systems, building on insights about the fundamental nature of complex biological and technological networks drawn from three converging research themes.

1. With molecular biology's detailed description of components and growing attention to systems biology the organizational principles of biological networks are becoming increasingly apparent.
2. Advanced technology's complexity is now approaching biology's. While the components differ, there is striking convergence at the network level of architecture and the role of layering, protocols, and feedback control in structuring complex multiscale modularity. New theories of the Internet and related networking technologies have led to test and deployment of new protocols for high performance networking www.hot.caltech.edu, www.netlab.caltech.edu
3. A new mathematical framework for the study of complex networks suggests that this apparent network-level evolutionary convergence within/between biology/technology is not accidental, but follows necessarily from the universal system requirements to be efficient, adaptive, evolvable, and robust to perturbations in their environment and component parts. www.cds.caltech.edu/sostools, www.cds.caltech.edu/~doyle

Biological systems are robust and evolvable in the face of even large changes in environment and system components, yet can be extremely fragile to small perturbations. Such universally robust yet fragile (RYF) complexity is found wherever we look. The amazing evolution of microbes into humans (robustness of lineages on long timescales) is punctuated by mass extinctions (extreme fragility). Diabetes, obesity, cancer, and autoimmune diseases are side-effects of biological control and compensatory mechanisms so robust as to normally go unnoticed.

RYF complexity is not confined to biology. The complexity of technology is exploding around us, but in ways that remain largely hidden. Modern institutions and technologies facilitate robustness and accelerate evolution, but enable catastrophes on a scale unimaginable without them (from network and market crashes to war, epidemics, and climate change). Understanding RYF means understanding architecture - the most universal, high-level, persistent elements of organization - and protocols. Protocols define how diverse modules interact, and architecture defines how sets of protocols are organized.

Chaos, fractals, random graphs, criticality and power laws inspire a popular view of complexity where behaviors that are typically unpredictable and fragile 'emerge' from simple interconnections among like components. But applied to the study of highly evolved systems, this attractively simple view has led to widespread errors and confusion. We will focus on a different, more rewarding take on complexity that derives from a revolution in the theory of distributed control driven by network technologies. It focuses on organization, protocols and architecture, and includes the 'emergent' as an extreme special case within a much richer dynamical perspective.

The Internet is an obvious example of how a protocol-based architecture facilitates evolution and robustness. The hourglass protocol 'stack' has a thin, hidden 'waist' of universally shared feedback control (TCP/IP) between the visible upper (application software) and lower hardware) layers. This allows 'plug-and-play' between modules that obey shared protocols; any set of applications that 'talks' TCP can run transparently and robustly on any set of hardware that 'talks' IP, accelerating the evolution of TCP/IP-based networks.

Similarly, in the microbial biosphere, genes that 'talk' transcription and translation protocols can move by horizontal gene transfer (HGT), also accelerating evolution a `bacterial internet.' The newly acquired proteins work better when they can use additional shared protocols such as group transfers and carriers in metabolism. Thus selection acting at the protocol level could evolve and preserve shared architecture, essentially 'evolving evolvability'.

All life and advanced technologies rely on protocol-based architectures. The evolvability of microbes and IP-based networks illustrate how dramatic, novel, dynamic changes on all scales of time and space can also be coherent, responsive, functional and adaptive, despite implementations that are largely decentralized and asynchronous. New genes and pathways, laptops and applications, even whole networks, can plug-and-play, as long as they obey protocols. Biologists can even swap gene sequences over the Internet in a kind of synthetic HGT.

Typical behavior is fine-tuned with this elaborate control and thus appears boringly robust despite large internal and external perturbations. As a result, complexity and fragility are largely hidden, often revealed only by catastrophic failures. Since components come and go, control systems that reallocate network resources easily confer robustness to outright failures, whereas violations of protocols by even small random rewiring can be catastrophic. So programmed cell (or component) 'death' is a common strategy to prevent local failures from cascading system-wide.

The greatest fragility stemming from a reliance on protocols is that standardized interfaces and building blocks can be easily hijacked. So that which enables HGT, the web and email also aids viruses and other parasites. Large structured rearrangements can be tolerated, while small random or targeted changes that subtly violate protocols can be disastrous. By contrast, the popular notion of complexity, continually recycled in series of 'new sciences', is one of nominal fragility in random interconnections of like components. Modeling and analysis both simplify because tuning, structure, and details are minimized, as is environmental uncertainty; and superficial patterns in ensemble averages (not protocols) define 'modularity'. An unfortunate 'clash of cultures' arises because architecture-based RYF complexity is utterly bewildering when viewed from this perspective. But the search for a 'deep simplicity and unity' remains a common goal.

Fortunately, our growing need for robust, evolvable technological networks means the tools for engineering architectures and protocols are becoming more unified, scalable and accessible. These will bring much-needed rigor and relevance to the study of complexity generally, including biology, but will not eliminate the need for attention to details relevant to particular domains and their components. Quite the contrary: both architectures and theories to study them are most successful when they facilitate rather than ignore the inclusion of of domain-specific details and expertise.

From brain to neurons to molecules and back again: circular causality in the organization of embodied cognition   Dr. Walter Freeman

Abstract: In my talk I address the question: how might concepts from complexity theory be used to simplify the modeling of brain structure and dynamics for knowledge-based engineering? Engineers have often searched for novel approaches to machine intelligence by asking how brains work. This search has become surreal, because most engineers get their understanding of brain properties secondarily from reports by neuroscientists, while those neuroscientists whose work they can profitably read have adopted their hypotheses and experimental tools from engineers. This circularity has trapped both engineers and neuroscientists in a hall of mirrors. While it is well established that brains are not computers, it is equally clear that brains are dynamical systems of a different kind, but in order to construct really new kinds of devices, we must escape the trap. Yet we face continuing sources of confusion, because we have to use computers to solve our equations and build new devices. The problem is not merely that engineers mistakenly categorize a neuron as a transistor and an action potential as a binary digit; it is that the computational metaphor is so pervasive that to reject it may seem perverse and obfuscate.

My way out of this trap is to review the history and philosophy of how and why the computational model for brains has become so entrenched. I use that background to return to fundamentals and derive a biological model of brain functions in terms of nonequilibrium thermodynamics. I rely on three organizing principles:

• Brains, like computers, are open systems, but brains use bodies to exchange energy and information by engaging their environments. Moreover, they are closed systems with respect to knowledge and meaning in cognition and experiential learning. An appropriate model is the intentional robot. www.scholarpedia.org/article/Intentionality.
• The underlying topology of vertebrate brains is random and scale-free; structured connections such as those in local networks, topographic maps, hubs, etc. evolve as departures from randomness by genetic evolution and experiential learning through the action-perception cycle. A new starting point is random graph theory and neuropercolation. www.scholarpedia.org/article/Scale-Free_Neocortical_Dynamics.
• State variables in brain models are determined by observers' methods of measurement, not by the brains in their continuum of physical and chemical aspects. The multiple space-time scales that we require for measurement with chemical assays, microelectrodes, imaging devices, etc., give networks of patches that have to be stitched together using the concept from synergetics of circular causality: entities at each level order the lower-level elements that create them. www.scholarpedia.org/article/Hilbert_transform_for brain_waves . A well-known acronym 'GOFAI' means 'good old-fashioned artificial intelligence'.

Dynamic Complexity in System of Systems - Ronald Johnson

Abstract: Levels of integration demanded by military operations as well as commercial products requires a robust engineering approach to manage dynamic complexity. A system of systems environment represents very high levels of complexity due to the changing nature of the interactions and the many ways each system may be employed. Applying a static, hierarchical approach to such an environment is insufficient in describing the evolving nature of a system of systems. An architecture centric methodology and understanding levels of interoperability required are important elements in achieving this integration.