Here is the strange rub of such a deep learning system: It learns, but it cannot tell us why it has learned; it assigns probabilities, but it cannot easily express the reasoning behind the assignment.
Our new formalism draws on utility-based approaches and expresses the reasoning in the form of arguments and objections, to enable full integration with value-based practical reasoning.
The calculus is high level, rather than minimal, and is consequently convenient for expressing and reasoning about software components, and different styles of composition.
That could probably be performed by observing you 24 hours a day but, since mind reading doesn't exist yet, you would also need to be expressing your reasoning out loud so the machine could learn to think like you.
Su Shi famously expressed this reasoning when he wrote that Du Fu was "preeminent ... because ... through all his vicissitudes, he never for the space of a meal forgot his sovereign".
I am sure that this is the reasoning expressed by some colleagues.
Ponting's reasoning, expressed to Don Argus in 2011, was no less logical than Waugh's: the captain is "the only person" who is ultimately "accountable" for the performance of the Australian cricket team, therefore, he must be a selector so that "he can … be truly accountable for what happens in his team".
The theoretical aspect is also expressed as articulate reasoning about the goal.
Then, the research focus was mainly on the cognitive processing styles (structured/discovery) and peoples' capabilities expressed as arithmetic reasoning (problem-solving) abilities.
Researchers were mainly interested in the machine-dimension of the HCI (Anonymous 2008) examining peoples' capabilities expressed as arithmetic reasoning (problem-solving) abilities (Bagley, 1991) to manage the computer operations.
