Inference Quotes - page 3
Traditional statistics is strong in devising ways of describing data and inferring distributional parameters from sample. Causal inference requires two additional ingredients: a science-friendly language for articulating causal knowledge, and a mathematical machinery for processing that knowledge, combining it with data and drawing new causal conclusions about a phenomenon.
Judea Pearl
Ambedkar traces Buddha's rational approach, which he values so much, to Kapila, the founder of the Samkhya-Darshana, the 'viewpoint' focusing on cosmology: 'Among the ancient philosophers of India the most preeminent was Kapila (') The tenets of his philosophy were of a startling nature. Truth must be supported by proof. This is the first tenet of the Samkhya system. There is no truth without proof. For purposes of proving the truth Kapila allowed only two means of proof-1) perception, and 2) inference.
Kapila
Is man ever a creature to be trusted with wholly irresponsible power? And does not the slave system, by denying the slave all legal right of testimony, make every individual owner an irresponsible despot? Can anybody fall to make the inference what the practical result will be?
Harriet Beecher Stowe
[T]ruly grand and powerful theories [...] do not and cannot rest upon single observations. Evolution is an inference from thousands of independent sources, the only conceptual structure that can make unified sense of all this disparate information. The failure of a particular claim usually records a local error, not the bankruptcy of a central theory. [...] If I mistakenly identify your father's brother as your own dad, you don't become genealogically rootless and created de novo.
Stephen Jay Gould
[H]istorical science is not worse, more restricted, or less capable of achieving firm conclusions because experiment, prediction, and subsumption under invariant laws of nature do not represent its usual working methods. The sciences of history use a different mode of explanation, rooted in the comparative and observational richness in our data. We cannot see a past event directly, but science is usually based on inference, not unvarnished observation (you don't see electrons, gravity, or black holes either).
Stephen Jay Gould
[T]ruly grand and powerful theories [...] do not and cannot rest upon single observations. Evolution is an inference from thousands of independent sources, the only conceptual structure that can make unified sense of all this disparate information. The failure of a particular claim usually records a local error, not the bankruptcy of a central theory. [...] If I mistakenly identify your father's brother as your own dad, you don't become genealogically rootless and created de novo. You still have a father; we just haven't located him properly.
Stephen Jay Gould
Let there be, not merely an indefinite succession, but a continuous flow of inference through a finite time; and the result will be a mediate objective consciousness of the whole time in the last moment. In this last moment, the whole series will be recognized, or known as known before.
Charles Sanders Peirce
In drawing an inference or conclusion from facts proved, regard must always be had to the nature of the particular case, and the facility that appears to be afforded, either of explanation or contradiction. No person is to be required to explain or contradict, until enough has been proved to warrant a reasonable and just conclusion against him, in the absence of explanation or contradiction.
Charles Abbott, 1st Baron Tenterden
Holists, either irredundant or redundant commit mistakes due to simple over-generalization (for example, that (β) always implies "Bushy Tail" which is true for only some subspecies) or systemic over-generalization to render the classification scheme more rational or symmetrical than it actually is (for example, falsely naming a subspecies "Bit QL" on the evidence that Q stands for "4-legged" and L stands for "Linear" together with the valid inference that a 4-legged linear creature exists.
A serialst is prone to list the subspecies by examining picture cards in Class A. If he is to be successful, he checks the relevance of the information entering his list by forming single predicate hypotheses.
Gordon Pask