BMM banner
Volume 1, Number 1
ISSN 2166-9732

Hover Button Hover Button Hover Button Hover Button Hover Button Hover Button


Table of Contents

    Front Cover              0

    Editorial: Start a Different Kind of Magazine              1
    by Juyang Weng

    Brain-Mind Institute: For Future Leaders of Brain-Mind Research              2

    Open Letter to the US President Obama: Is the US Foreign Policy Scientifically Shortsighted? banner             3 - 4
    by Juyang Weng
    Abstract: All minds are groupish and shortsighted in nature. The aftermath of Richard Nixon’s China visit demonstrated that a scientifically correct foreign policy is to make friends with foes, counter intuitive to many souls. Our brains blinded us. Scientific principles, e.g., checks-and-balances of government power, seem more convincing and effective in converting foes than shallow and ideological slogans like “human rights”.

    Brain Stories 1: Naivety in Everybody banner             5 - 6
    by Brian N. Huang
    Abstract: Meant for layman readers, this series uses real world stories to explain how a single brain works computationally inside its skull and how multiple brains work together to give rise to group intelligence. Hopefully, this series is useful for us humans to see the weakness of our current governing systems, in developed countries and developing countries alike, and how a country can develop earlier and better. It also explains some key mechanisms to make a robot learn skills that its human programmer does not have. This installment is about naivety, in childhood and adulthood; in you and in your officials. Brains are naive for various tasks, making strategic errors with high costs. This installment raises a few nation-scale naiveties to be discussed in future installments of this series.
    Index terms: brain, mind, law, group intelligence

    From Problem Solving to Problem Posing banner             7 - 8
    by Yoonsuck Choe and Timothy A. Mann
    Abstract: Artificial intelligence and machine learning approaches are both very good at problem solving. However, the various methods accumulated in these fields have not been able to give us truly autonomous agents. The main shortcoming is that the problems themselves are formulated by human designers and subsequently fed to the problem solving or learning algorithms. The algorithms do not question the validity of the problems nor do they formulate new problems. This latter task is called “problem posing”, and is in fact an active area in education research. In this article, we will discuss the importance and relevance of problem posing to autonomous intelligence and speculate on key ingredients for effective problem posing in an AI and machine learning context.
    Index terms: problem posing, education, artificial intelligence

    Why Should the CVPR Community See that Output Is Not Only Output? banner             9 - 10
    by Christopher S. Masfis
    Abstract: The currently prevailing methods in the computer vision and pattern recognition (CVPR) community require images for system training. Many of such methods require manually drawn object contours to segment the pixels from each object of interest from those pixels in other parts of the image. Does a human child require such object-contour annotation to learn how to detect, recognize, and segment objects from cluttered natural scenes? Weng argued for a negative answer. He explained that a brain autonomously learns directly from its physical environment using not only non-annotated continuous video of dynamic and cluttered scenes, but also its video-synchronized actions — output is also input. Body-environment interactions give rise to brain representations. In particular, contour annotation is not necessary for a brain, neither for a machine.
    Index terms: pattern recognition, computer vision, brain

    Back Cover              11

Established since June 2012