Extended Summary of "Living Things Are Not (20th Century) Machines : Updating Mechanism Metaphors in Light of the New Science of Machine Behavior"
- 1. EXTENDED SUMMARY OF “LIVING THINGS ARE NOT (20TH CENTURY) MACHINES: UPDATING MECHANISM METAPHORS IN LIGHT OF THE NEW SCIENCE OF MACHINE BEHAVIOR” Bongard J and Levin M (2021) Living Things Are Not (20th Century) Machines: Updating Mechanism Metaphors in Light of the Modern Science of Machine Behavior. Front. Ecol. Evol. 9:650726. doi: 10.3389/fevo.2021.650726 Laureando: Simone Cappiello Relatore: Prof. Eric Medvet Anno Accademico 2021/2022
- 2. QUESTIONS AND PURPOSES OF THE PAPER What is meant by “machine”? Need to update the “life as machine” metaphor Improving definitions so that they can also apply to biology Rise of a new multidisciplinary field mixing aspects of information sciences, physics and biology.
- 3. MACHINE/LIFE DUALISM: REDRAWING THE BOUNDARIES MACHINES LIFE ARGUMENTS Independent Interdependent Machines that perform communication and cooperation with the environment in which they are placed exist. Designed Evolved Evolutionary algorithms proved that robots, synthetic organisms, and other machines can be evolved.
- 4. MACHINE/LIFE DUALISM: REDRAWING THE BOUNDARIES MACHINES LIFE ARGUMENTS Linearly Modular Hierarchical and self- similar Still a consistent difference, though efforts to implement self-similarity in machines will increase. Not capable of intelligence Capable of intelligence Debates about what intelligence is are still open. Biohybrid systems where machines influence behavior and intelligence exist.
- 5. MACHINE/LIFE DUALISM: REDRAWING THE BOUNDARIES MACHINES LIFE ARGUMENTS Can be studied in a reductionist framework Cannot be studied in a reductionist framework Technologies like neural networks and swarm AI need a top-down approach to be understood. Have clear hardware/software distinctions Doesn’t have clear hardware/software distinctions The distinction is blurry in both biology and technology.
- 6. IMPROVING DEFINITIONS MACHINE System using principles of physics and computation to amplify the power of an agent to make changes in its environment. Not necessarily a physical entity. ROBOT Machine affecting its environment through physical action.
- 7. IMPROVING DEFINITIONS PROGRAM Abstract procedure that can be executed on many different physical systems that support computation. Not necessarily written by humans and linear one-step-at- a-time instructions. Concept extended to biological systems. SOFTWARE/ HARDWARE Hardware is usually harder to modify than software. Though, taking for granted a sharp distinction between the two might not be useful in creating intelligent machines and studying biological adaptation.
- 8. HYBRIDIZATION: Classifying hybrid systems • FIGURE 1 (Source: Bongard and Levin 2021). Hybrid systems placed on a multi- scale axis based on the level of organization (from cells, to individuals, to swarms, z axis), degree of autonomy (from merely mechanical to high-level cognition, y axis ) and degree of design (from designed to evolved systems, x axis)
- 9. MULTIDISCIPLINARY BENEFITS OF A NEW SCIENCE OF MACHINES Improving techniques of reverse-engineering and multi scale analysis, useful in regenerative medicine and developmental biology. Improving understanding of cognition. Using biomimicry in artificial intelligence research.
- 10. CONCLUSION Biology and Computer Science Science fields dealing with similar issues in different media A better cooperation between the two will open new perspectives in empirical research and widen the space of possible embodied computing systems.