時間：2015年1月30日（
地點：本所會議室(第二教學大樓2樓人社2
論文題目：「什麼是客觀感知世界的條件
Tuesday, January 27, 2015
Thursday, January 15, 2015
Wednesday, January 07, 2015
Dr. Guillaume Rochefort-Maranda will come to our institute and give a talk.
Theme:"Simplicity
Matters?
The
Case
of
Non-parametric
Models"
Date & Time: Thursday, 7 Jan., 1:30-3:30 p.m.
Venue: Meeting Room, Institute of Philosophy of mind and cognition, NYMU
About the theme
In this presentation, I claim that influential arguments concerning the importance of parametric simplicity for model selection have been biased by their focus on parametric models. See (Foster & Sober 1994) (Foster 2001) (Sober & Hitchcock 2004). Such a focus leads us to believe that there is a fundamental trade‐off between parametric simplicity and goodness of fit. But no such trade‐off is considered when we select non‐parametric models, like KNN1 regression models. We can increase the fit of KNN models by keeping the number of adjustable parameters to 1.
This leads me to point out that the important trade‐off that is made as we select any kind of model is between the bias and the variance of an estimator for a dependent variable. Consequently, I explain why a favored selection criterion for the proponents of parametric simplicity, i.e., the AIC (Akaike Information Criterion) is not optimal (in any scenario) in order to tell if we have made a reasonable bias/variance trade‐off. A selection criterion based on crossvalidation is more appropriate.
Key References
Forster, M. (2001), “The New Science of Simplicity”, (In A. Zellner, H.
Keuzenkamp, and M. McAleer (Eds.), Simplicity, Inference and Modelling. (pp. 83‐
119). Cambridge: Cambridge University Press)
Forster M. and Sober, E. (1994), “How to Tell When Simpler, More Unified, or
Less Ad Hoc Theories will Provide More Accurate Predictions”, The British
Journal for the Philosophy of Science, 45: 1 ‐ 35.
Hitchcock, C. and Sober, E. (2004) “Prediction Versus Accommodation and the
Risk of Overfitting”, The British Journal for the Philosophy of Science, 55: 1‐34
1 KNN
About the speaker
Guillaume Rochefort-Maranda, Laval University, Québec, Canada
AREAS OF SPECIALISATION
Philosophy of Science
Epistemology
Philosophy of Statistics and Probability
AREAS OF COMPETENCE
Philosophy of Mind
Philosophy of Language
Philosophical Logic
Metaphysics
Applied Ethics
Big Data Studies
Wednesday, December 10, 2014
FALL 2014 Lectures Series (No.8):Aristotle’s view of causation
Date & Time: Friday, 12 Dec., 2-4 p.m.
Venue: Rm 202, Centre for Humanities and Social Sciences Education, NYMU
About the theme
Aristotle
provides a compelling account of the metaphysics of powers but focuses little
on how they act in combination[1].
Contemporary science, by contrast, is centrally concerned with nomological machines (often termed mechanisms): arrangements of features
and powers whose repeat operation can rise to laws of nature. Nomological
machines typically involve the simultaneous exercising of multiple powers, which
we may take to be Aristotelian[2],
so that the question of how such powers combine is central to modern science –
but is unfortunately far from adequately answered.
The
exercisings of some powers seem to be contributions to immediate changing, e.g.
powers which result in forces, and perhaps (but perhaps not) the powers to heat
/ cool, to dissolve, or to chemically react. The exercisings of other powers
are processes which take time, e.g. the power of a pendulum to swing, a cistern
to produce a flush, a neuron to fire, or a glass to smash. I explore a range of
examples featuring combinations of powers with each of these two sorts of
timing: this suggests a diversity of case types of how powers combine (which
should perhaps not surprise an adherent of the Stanford School). However, it
seems that (often at least) the exercise of a power that occurs over time
relies on structures within the machine to coordinate the relevant powers of
the parts. For such time extended powers, it seems that we might reframe the
question of how the exercising of powers combine, as a question concerning how
the structures which license those powers combine in forming the machine
arrangement. I explain how this approach might lead to a more unified account
of how Aristotelian powers combine.
About the speaker
John Pemberton, Associate at the Centre for Philosophy of the Natural and Social Sciences at LSE
專長：
Change, powers, causation, arrangement, structure, laws
FALL 2014 Lectures Series (No.7):Physical vs. mathematical structures: is there a difference?
Date & Time: Wed, 10 Dec., 2-4 p.m.
Venue: Rm 242, Centre for Humanities and Social Sciences Education, NYMU
About the speaker
Alexandre Guay, 比利時 魯汶大學副教授，物理學哲學家
專長：
Philosophy of Physics, Ontology, Metaphysics, Philosophy of Biology, History of Science, Philosophy of Science, Evolutionary Biology, Aristotle, Behavioral Ecology
Subscribe to:
Posts (Atom)