Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
31 Cards in this Set
- Front
- Back
Modus Ponens |
If P implies q and q is true, then p is also true |
|
Modus tollens |
If p implies q and q is false, then p is false |
|
Bacon's Idols of the tribe |
Tendency of humans to see patterns where none exis |
|
Bacon's Idols of the Cave |
Errors caused by personal biases derived from one's culture, upbringing, or personal tastes. |
|
Bacon's Idols of the Forum |
Imprecision and disagreement in the definition of natural language terms |
|
Bacon's Idols of the Theatre |
False systems of reasoning (logic errors) |
|
Method of Agreement |
Co-occurrence of factors {ABCD} occurs w/ {wxy} {AEFG} occurs w/ {wtuv} |
|
Method of difference |
If we take away the expected cause, does the expected response disappear? |
|
Joint method of difference and agreement |
Method of agreement "a sufficient condition" Method of difference "a necessary condition" {ABC} occur w/ {xyz} {ADE} occur w/ {xvw} {BC} occur with {y,z} therefore A ==> x V (x ==> A) |
|
Method of Concomitant Variation |
Same as joint, but with varying amounts |
|
Method of Residues |
{ABC} occur w/ {xyz} B ==> y C ==> z therfore A ==> x |
|
Inductive v. Deductive |
Inductive does not guarantee the truth of the conclusion, deductive reasoning cannot provide information that was not already encoded in the premises and rules of logic. |
|
Telford: Fundamental principles of experimental design |
Randomization, Replication, Blocking, Orthogonality, Factorial experimentation |
|
What are treatments |
The different procedures we want to compare |
|
What are Experimental Units |
The things we're applying treatments to |
|
What are Responses |
The outcomes of applying treatments to experimental units |
|
What are Measurement Units |
Actual objects on which the response is measured |
|
What is inductive inference? |
Looking at probability as a fraction of possible events |
|
What is Ddeductive Inference? |
Includes everything we know and uses Baysian logic to test |
|
Solomon four-group design |
1. divided into 4 groups randomly (two control, two experimental) 2. One set is given pre- and post-tests, the other only postRepli |
|
Risidual |
Distance between observed and actual values Risidual sum of squares: summation of all em (the least squares criterion) |
|
Replicates? |
Should vary ALL confounding factors |
|
R^2 |
1 - RSS/variance aka - the proportion of the variance explained by the model |
|
Variance Reduction Techniques |
Common Random Numbers Conditional Expectation |
|
Performance Metric |
Dimension of an algorithm performance that can be measured (time, space, precision) |
|
Parameter |
Any property that affects the value of a performance indicator |
|
Performance Indicator |
A quantity associated with a performance metric |
|
Experimental Design Template |
- Question - Performance indicators - Factors - Levels - Trials per design point - Design points - Outputs |
|
TSS |
sum(y - yavg)^2 = factor1 + ... + factork + SSE |
|
Greater SST? |
Larger(SST sum of squares for treatment) give greater the evidence for rejecting the null hypothesis |
|
Sums of Squares Treatment |
n sum(mean of treatment - mean of system)^2 |