Skip to main content
main-content

Über dieses Buch

This book attempts to connect artificial intelligence to primitive intelligence. It explores the idea that a genuinely intelligent computer will be able to interact naturally with humans. To form this bridge, computers need the ability to recognize, understand and even have instincts similar to humans.

The author organizes the book into three parts. He starts by describing primitive problem-solving, discussing topics like default mode, learning, tool-making, pheromones and foraging. Part two then explores behavioral models of instinctive cognition by looking at the perception of motion and event patterns, appearance and gesture, behavioral dynamics, figurative thinking, and creativity. The book concludes by exploring instinctive computing in modern cybernetics, including models of self-awareness, stealth, visual privacy, navigation, autonomy, and survivability.

Instinctive Computing reflects upon systematic thinking for designing cyber-physical systems and it would be a stimulating reading for those who are interested in artificial intelligence, cybernetics, ethology, human-computer interaction, data science, computer science, security and privacy, social media, or autonomous robots.

Inhaltsverzeichnis

Frontmatter

Primitive Problem-Solving

Frontmatter

Chapter 1. Introduction

Abstract
We watch a mobile robot making his way to clean the floor in a furnished room. He starts moving in an outward spiral. Then, he bumps into the wall or a piece of furniture, and he bounces back in a random direction. After a while, he has covered all of the areas that he can reach. He is blind, without a camera. However, he can touch a surface to feel whether it is a wall or the stairs. When he feels hungry, he moves over to the power charger for a boost of energy. So as not to anthropomorphize his purposes, I used a long-exposure filming app on my phone to trace his path in a single image. The path is a laborious sequence of irregular, angular segments – not quite a random walk, nor optimal movement but complex and hard to describe. Everything seems spontaneous, but it works out eventually and the floor is cleaned. I then compare the traced path of this cleaning robot to that of a foraging ant, a stream of Internet packets, and even a malicious computer virus. I found that all of their paths are surprisingly similar: they have a general sense of purpose, but they cannot foresee all of the obstacles in between. Therefore, they must adapt their course repeatedly with allowable changes in direction and take detours around any barrier.
Yang Cai

Chapter 2. Default Mode

Abstract
We often take everyday common sense for granted, frequently acting without thinking. Many subconscious processes are instinctual, but they can also be learned and evolve. When a baby is born, its communication is limited to one mode – crying – which signals to the parents if the baby is feeling a lack of attention, hunger, or pain. When a creature has limited means of survival or is in risky situations that other means are not available, the creature typically reverts to a survival or backup state, so-called default mode, and performs survival actions. It uses simple algorithms to cope with complex and dynamic situations. In this chapter, we will explore how to use default mode algorithms and default knowledge to solve challenging problems in modern control, network, and vision systems such as collision handling in networks, exception-handling in autonomous vehicles, and object recognition.
Yang Cai

Chapter 3. Pheromone Trails

Abstract
Pheromones in nature are chemical messages that act within a species. It was Karlson and Lüscher who coined the term “pheromone” to cover the wide range of substances secreted to the outside world by individual organisms and received by other individuals of the same species. Pheromones are used widely by insects. The analogy between this mode of communication within an insect society and communication within the body by means of hormones, led to these substances also being referred to as “social hormones.” (Wigglesworth VB (1970) Insect Hormones. pp. 134–pp. 141. W.H. Freeman and Company.) These chemical messages have diverse biological effects and differ widely in their modes of action. In practice, the term “pheromone” proves useful in describing behaviors such as trail formation, defensive secretions, and social coherence.
Yang Cai

Chapter 4. Foraging Behaviors

Abstract
Foraging is an instinctual behavior for living. It is a continuous activity to search surrounding areas for food either passively or actively. This requires the actions of hunting and gathering. Foraging is a broad search over an area in order to obtain something, whether it is ants looking for green leaves, bees scouting for flower pollen, or even people seeking specific information. Foraging describes how animals behave to satisfy their life-sustaining needs. In the cyberspace, foraging means to actively seek, gather, share, and consume information.
Yang Cai

Chapter 5. Primitive Learning

Abstract
Natural selection has produced instinctive behaviors through the slow and gradual accumulation of numerous slight, yet advantageous variations. The result is automation, a behavior that is second nature. This is called the Baldwin Effect. This principle can be applied to both humans and machines. In order to enable a machine to learn as efficiently as a human or animal, we need to understand primitive learning processes within humans, animals, and even insects. In this chapter, we explore primitive learning behaviors, including perceptual adaptive learning, peak-shift, lateral learning, learn-by-tapping, and indirect learning from virtual experiences, such as playing or gaming.
Yang Cai

Chapter 6. Tool Making

Abstract
One quality that defines human instinct is not our compulsion to survive, but our ability to create things in the most unpredictable ways. Hacking or improvisation, in a technical context, is to solve a problem with the tools and materials immediately at hand, in ways the materials were not originally designed for. It is problem solving on-the-fly in spontaneous moments of sudden inventiveness.
Yang Cai

Instinctive Cognition

Frontmatter

Chapter 7. Perceiving Motion Patterns

Abstract
Humans are visual explorers. Human vision is a highly dynamic process. Consider our instinctive reaction to every sudden change. Each change may be inconsequential; but it may also be vitally important. We inspect the world of objects and events with rapid eye movements, which consist of frequent gaze shifts at two to four fixations per second. During a fixation period, the underlying neural processes are updated in less than half a second (Öğman H and Breitmeyer BG, eds. (2006) The First Half Second. The MIT Press). Subconscious motion perception enables us to perceive motion patterns, correlated or anomalous events in our surrounding environment. Human vision is by far the best computer in our everyday life. In this chapter, we explore how to map spatiotemporal data into motion pictures, and how to incorporate human perceptions in pattern discovery in motion and events.
Yang Cai

Chapter 8. Sensuality

Abstract
The word sensuality comes from the root word sense, which pertains to our five senses: sound, sight, touch, smell, and taste. The Greek word often means a total devotion to the pleasures of the physical senses, sometimes implying outrageous conduct. Sensuality is different from sexuality. Sensuality is a much broader term that goes beyond sexual attraction and eroticism. Sensuality is also subtler and more cognitive than sensation, which is a reflexive feeling or reaction resulting from excitement. In this chapter, we explore how to incorporate sensuality into humanoid robot design, how to analyze the sensuality in human voices, and how to transform a computerized or robotic monotone voice into a sensual voice. We also explore how to detect and analyze sensual body shapes and gaits.
Yang Cai

Chapter 9. Euphoria Dynamics

Abstract
Why do we get hungry and then feel satiated when we eat? How do these biological compulsions arise? Our thoughts, feelings, and bodily sensations correlate with the activation of specific parts of our neural circuitry. Such a dynamic process happens by means of neurochemical changes in our brain that keep us alive and reproducing. In this chapter, we explore a mathematical and electronic model of euphoria dynamics, and the potential for a computer to understand, or even to have such a behavior.
Yang Cai

Chapter 10. Describing Faces

Abstract
Over the last several decades, facial recognition technology has become increasingly accurate. However, despite advances in technology, these systems are not yet as good as what was envisioned in crime scene investigation (CSI) films. For example, facial recognition technology failed in the Boston Marathon bombing manhunt in 2013 (Gallagher S (2013) Why facial recognition tech failed in the Boston bombing manhunt. http://​arstechnica.​com/​information-technology/​2013/​05/​why-facial-recognition-tech-failed-in-the-boston-bombing-manhunt/​). The two bombers, Dzhokhar and Tamerlan Tsarnaev, were both in the facial image database. There were photos of the suspects, but the system could not find a match, or at least could not come up with one before the suspects were identified by humans. Under the best circumstances, facial recognition can be extremely accurate, returning the right person as a potential match within ideal conditions, e.g. front-view faces where all photos are shot from the same angle and with the same lighting. To reach that level of accuracy in real-world footage, which is often blurry, with different poses and lighting, computers almost always require a degree of skilled human guidance. According to the NIST report on the evaluation of 2D still-image facial recognition algorithms, the facial recognition accuracy rate decreases linearly with the logarithm of the population size of the image database. In all cases, human adjudication is ultimately necessary for verification (Grother P, Quinn GW, and Philips PJ (2011) Report on the evaluation of 2D still-image face recognition algorithms. NIST Interagency Report 7709, August 24, 2011).
Yang Cai

Chapter 11. Figurative Thinking

Abstract
Charles Darwin believed that language was half art, half instinct. Languages originated from sharing survival skills in ancient times. In the Stone Age, people learned to make tools through observation, mimicry, and gestures, eventually developing verbal languages. What makes human languages unique are the written languages that have been passed down over time. Figurative thinking is our instinct. It follows the principle of least effort as George Zipf predicted for words. Future intelligent systems are moving toward more figurative thinking with more figurative contents and interactions. In this Chapter, we explore the origin of figurative languages, digital encoding methods, and applications in passcode, programming, and social media.
Yang Cai

Chapter 12. Machine Creativity

Abstract
Creativity is the ability to make new things or think of new ideas. Creativity has been mystified as an act of the divine or genius in art, literature, music and invention. But in Nature, all species must have some kind of serendipity, or so-called “blind creativity,” in order to survive (Mueller ET and Dyer MG (1985) Daydreaming in human and computers. Proceedings of the ninth international joint conference on artificial intelligence, Los Angeles, CA, August 18–24, 1985). As we discussed before, insects have a default mode where they demonstrate spontaneous alternation behaviors to avoid deadlocks, bias, and blind spots. Living cells maintain a delicate balance between inheritance and mutation in order to evolve; insects explore different foraging paths around a new home; even the insect-like floor cleaning robot Roomba takes different routes whenever it bumps into a wall in order to cover the entire floor. Creativity by nature is an instinctive survival behavior. The difference between a repetitive routine and creative thinking lies in the injection of some randomness. The randomness must be organized to be efficient. Creativity is controlled chaos. In this Chapter, we explore where do creative ideas come from? How do ideas crossover, mutate and develop over time? What are major challenges in machine creativity?
Yang Cai

Evolving Cybernetics

Frontmatter

Chapter 13. Self-Awareness

Abstract
According to ethnologists, humans are simply survival machines programmed to preserve and replicate our genes. In order to secure this process, we have developed self-awareness to identify ourselves and to avoid life-threatening situations at multiple levels, from our cells to our existence as organisms. In this Chapter, we explore self-defense systems in nature and cyberspace, the behavioral models of malicious programs, biomorphic defense models, and collective consciousness in the Internet of Things.
Yang Cai

Chapter 14. Stealth

Abstract
When we talk about hiding information, we often refer to data encryption. It is a mathematical way to convert signals to seemingly random noise. Only the authorized receivers have the key to unlock the data and read the signals. However, encryption itself also reveals a signal – the file contains some secrets that the sender wants to hide. Once the suspicious package is located, analysts may use a powerful computer to run decryption algorithms to unlock the secret data.
Yang Cai

Chapter 15. Visual Privacy

Abstract
Our sense of privacy comes from cavemen because we are still the hunter-gatherers of 100,000 years ago. We always feel uncomfortable if someone is staring at us because we are hunters and we don’t want to be hunted. Privacy is also about territorial rights. We take private space as a measure of security. A private space is normally the most vulnerable place. Anything that enters such a space is seen as a threat or a discomfort, triggering emotional, even aggressive reactions. In this Chapter, we discuss visual privacy for cavemen and cavewomen in the digital age, including personal territories in physical and cyber spaces, privacy technologies, and privacy research methodology.
Yang Cai

Chapter 16. Navigating and Covering

Abstract
Searching, foraging, and cleaning tasks often combine navigation and covering. The goal is to cover the desirable area with minimal traveling time. In this Chapter, we explore primitive navigating and covering heuristics such as wall following, and the Hansel and Gretel method. We apply them to solve the tessellation problem, or called traveling workstation problem which combines navigating and covering simultaneously.
Yang Cai

Chapter 17. Autonomy

Abstract
We tend to view a robot as a servant. But what if the robot had its own mind and could act on its own? We have an implicit fear of losing control to more powerful creatures, such as autonomous robots. According to Stephen Hawking, the primitive forms of artificial intelligence developed so far have already proven very useful. However, Hawking fears that the development of full artificial intelligence could surpass human intelligence, and take off on its own and to be redesigned at an ever-increasing rate, reaching a so-called “singularity.” Humans, who are limited by slow biological evolution, could not possibly compete and would be superseded as Hawking suggested (Cellan-Jones R (2014) Stephen Hawking warns artificial intelligence could end mankind, BBC News, 2 December 2014 http://​www.​bbc.​com/​news/​technology-30290540). In this Chapter, we explore principle elements in autonomic system design and how to incorporate human instinct and simplicity in autonomous systems.
Yang Cai

Chapter 18. Survivability

Abstract
Survival is the most instinctual behavior of all animals. It is also the most essential design strategy used in robotic and network systems, which were originally built for indoor, stationary, and isolated use. These systems have become increasingly more outdoor, mobile, and networked in recent years. The greatest concerns of our generation are battery life, wireless signals, computer attacks, and manmade disasters – all relatively new instincts in our digital lives. In this chapter, we review the tasks, strategies, and algorithms for survival in physical and digital environments. Traditionally, design strategies for these two worlds are separate. However, they share many survival models such as cliff detection, cache, collision avoidance, self-healing, and improvisation. They have gradually merged into one, observable in cyber-physical systems as large as cities with connected vehicles, resilient infrastructures, power grids, and mobile devices.
Yang Cai

Backmatter

Weitere Informationen

Premium Partner

    Bildnachweise