/Write down the probability distribution of t given y. Use the…
Question Answered step-by-step /Write down the probability distribution of t given y. Use the… /Write down the probability distribution of t given y. Use the principle of maximum likelihood to derive an expression for the corresponding error function defined over a set of training data comprising input vectors xn and targets tn, where n = 1, . . . , N. Write down a suitable form for the output unit activation function y = g(a). Hence evaluate the derivative of ln P(t|y) with respect to a. [10 marks] 6 CST.98.9.7 9 Security Describe those provisions of the following Acts of Parliament that are relevant to computer security: • The Civil Evidence Act of 1968 • The Police and Criminal Evidence Act of 1984 • The Data Protection Act of 1984 • The Computer Misuse Act of 1990 [12 marks] A software house incorporates a time-lock which causes their product to stop working if it is not supplied with a suitable password every 6 months. What risks are they running? [4 marks] A hospital de-identifies patient records by removing names and addresses, leaving only the patient’s postcode and date of birth as an identifier. These records are then sold to researchers and drug companies. What risk is the hospital running? [4 marks]? Consider Shannon’s third theorem, the Channel Capacity Theorem, for a continuous communication channel having bandwidth W Hertz, perturbed by additive white Gaussian noise of power spectral density N0, and average transmitted power P. (a) Is there any limit to the capacity of such a channel if its signal-to-noise ratio P N0W can be arbitrarily increased? If so, what is that limit? [2 marks] (b) Is there any limit to the capacity of such a channel if, leaving N0 and P unchanged, its bandwidth W in Hertz can be arbitrarily increased? If so, what is that limit? [3 marks] [Hint: ln(1 + x) = x − x 2 2 + x + · · · where −1 < x < 1.] Explain why smoothing a signal, by low-pass filtering it before sampling it, can prevent aliasing. Draw a graph in the Fourier domain which illustrates the origin of aliasing, and also show in the picture how smoothing solves the problem. What would be the most effective low-pass filter to use for this purpose? Draw it in the Fourier domain. [10 marks] Suppose that women who live beyond the age of 70 outnumber men in the same age bracket by three to one. How much information, in bits, is gained by learning that a certain person who lives beyond 70 happens to be male? [5 marks] 9 [TURN OVER CST.98.9.10 12 Computer Vision It could be said that the central problem of pattern recognition is the relation between the within-class variability and the between-class variability for the patterns that one would like to recognise. Explain this problem in the case of face recognition, treating separately the problems of (a) face detection (distinguishing faces from non-faces) (b) face identification (c) face interpretation (classifying the expression and pose angle of the face) How do the forms of variability for faces influence each of the three tasks? Is within-class variability ever helpful, and between-class variability ever harmful, to the performance of the task? What role can statistical decision-theory play in formalising and solving these problems? [20 marks] 13 Types What is meant by the terms safety and type soundness in connection with programming languages and their type systems? [2 marks] Explain how a na¨ıve combination of ML-style polymorphism and references results in an unsound type system. Your account should include the axioms and rules of the type system, but need not give a formal definition of the operational semantics. [15 marks] State, without proof, how the rule for typing let-expressions can be restricted to restore type soundness. [3 marks] 10 CST.98.9.11 14 Numerical Analysis II Explain what is meant by local error and global error in methods for the solution of ordinary differential equations (ODEs). If a typical method has local error O(h 3 ), what would you expect the global error to be? [3 marks] Euler's method for solution of y 0 = f(x, y) can be expressed as yn+1 = yn + k1. From the Taylor series, find an expression for k1. [2 marks] The Runge-Kutta method RK2 is yn+1 = yn + 1 2 (k1 + k2) where k1 is the increment used by Euler's method, and k2 = h f(xn + h, yn + k1). In terms of Euler's method, what does the quantity k2 represent? [2 marks] Assume that RK2 is carried out with step sizes h and h/2, and that y(h) (xn+1) = y(xn+1) + Cnh 2 + O(h 3 ). Derive an estimate of the error En = |Cn|(h/2)2 in y(h/2)(xn+1). [3 marks] Let ε be the target error per unit step. Why, in step-size control for RK2, is ε 0 = ε/8 taken as the target error correspondingRKImage transcription text1) Overview of the question 2) Screenshot of schematic entry (explanation) 3) Specify your design input andoutput 4) PIN Assignment explanation 5) Screenshot of circuit implementation on the DE2-115 Board 6) Showyour work for each sub-question 1. Construct and test a Verilog module that implements the follo... Show more... Show more Computer Science Engineering & Technology C++ Programming BIOL 174 Share QuestionEmailCopy link Comments (0)
