Скачать 238.92 Kb.

TOPICS IN FINANCIAL MATHEMATICS
1. Financial markets, financial derivatives, payoff functions 2. Asset price model 3. BlackScholes analysis; American and European options; BlackScholes formula 4. Variations on BlackScholes models; Future options 5. Numerical methods: Monte Carlo method, binomial method, finite difference method, fast algorithms for solving linear systems 6. Exotic options 7. Pathdependent options: General method, average strike options, lookback options 8. Bonds and interest rate derivatives: Bond models, interest models APPROXIMATION THEORY
Best polynomial and rational approximation, moduli of smoothness, JacksonTiman type quantitative direct estimates for polynomial approximation, converse theorems, positive linear operators, Korovkin theorems, interpolation (Lagrange, Newton, Hermite), Fourier series. APPLIED NUMERICAL ANALYSIS
1. Approximations of differential equations (finite difference, finite element, Galerkin and collocation methods) 2. Applications (Heat transfer, Maxwell equations, airpollution transport model, torsion of noncircular sections, irrotational flows, BlackScholes' equation) 3. Operator splitting techniques, matrix exponentials and their applications to airpollution transport models and to Maxwell equations 4. Qualitative properties of mathematical and numerical models 5. Computer examples and implementations MATHEMATICAL MODELS IN BIOLOGY AND ECOLOGY
1. Discrete and continuous single species models. Exponential and logistic growth. The delayed logistic equation 2. Multispecies communities: competition, comensualism, coexistence. 3. Predatorprey models. The LotkaVolterra model and more complicated models (Gause, Kolmogorov). Preydependent and ratiodependent predation. 4. Chemical reaction kynetics: MichaelisMenten theory 5. Simple oscillatory reactions. Nerve impulses and HodgkinHuxley theory. FitzHughNagumo model. 6. Reactiondiffusion equations. Convection, advection. Chemotaxis. 7. Ecological epidemiology: integrated pest management strategies. 8. Numerical simulations and visualisations by means of XPP, Phaser, Maple (or equivalent) THE MATHEMATICAL THEORY OF INFECTIOUS DISEASE PROPAGATION
1. Basic concepts of mathematical epidemiology. Deterministic models. Compartmental models. 2. Single population models with constant population size. Models with no immunity. 3. Models with nonconstant population size and immunity effects. Basic reproduction number of a disease. Stability and persistence. 4. Infective periods of fixed length. Models with delay. Arbitrarily distributed infective periods. 5. Seasonality and periodicity. Orbital stability of periodic solutions. 6. Models with pulse vaccination. 7. Multigroup models (models with patchy structure). 8. Numerical simulations and visualisations by means of XPP, Phaser, Maple (or equivalent). EVOLUTIONARY GAME THEORY AND POPULATION DYNAMICS
1. Evolutionary stability. Normal form games. Evolutionarily stable strategies. Population games 2. Replicator dynamics. The equivalence of the replicator equation to the LotkaVolterra equation. The rockscissorspaper game. Partnetship games and gradients 3. Other game dynamics. Imitation dynamics. Monotone selection dynamics. Bestresponse dynamics. Adjustment dynamics. A universally cyclic game. 4. Adaptive dynamics. The repeated Prisoner's Dilemma. Adaptive dynamics and gradients. 5. Asymmetric games and replicator dynamics for them. 6. Population dynamics and game dynamics 7. Game dynamics for Mendelian populations 8. Numerical simulations and visualisations BIOINFORMATICS
Stochastic models: HMMs, SCFGs and timecontinuous Markov models and their algorithmic aspects.
To learn the stochastic transformational grammars, especially HMMs and SCFGs To learn timecontinuous Markov models describing sequence evolution To learn the algorithmic background of these models To learn the statistical background and tools, like Maximum Likelihood and Expectation Maximization
The students will be able to read and understand scientific papers related to the topic.
Lecture 1. Theory: Score based dynamic programming algorithms. Linear, concave and affine gap penalties. Lecture 2. Theory: Conditional probability, Bayes theorem. Unbiased, consistent estimations. Statistical testing. Local alignment, extreme value distributions for local alignments, p and E value estimations. Lecture 3. Theory: Hidden Markov Models. Parsing algorithms: Forward, Backward and Viterbi. Posterior probabilities. Expectation Maximization. The BaumWelch algorithm. Lecture 4. Theory: Profile HMMs. Aligning sequences via profileHMMs. PairHMMs. Practice: HMM topology design. Lecture 5. Theory: Substitution models. Felsenstein’s algorithm for fast likelihood calculation of a tree. Lecture 6. Theory: Predicting protein secondary structures with profile HMMs and evolutionary models. Gene prediction with HMMs. Lecture 7. Theory: Modeling insertions and deletions with timecontinuous Markov models: The ThorneKishinoFelsenstein models. Lecture 8. Theory: Describing the TKF models as pairHMMs. Extension to many sequences: multipleHMMs. The transducer theory for evolving sequences on an evolutionary tree. Lecture 9. Theory: Stochastic transformational grammars. Stochastic regular grammars are HMMs. Stochastic ContextFree Grammars. Parsing algorithms for SCFGs: Inside, Outside and CYK. Lecture 10. Theory: Posterior decoding of SCFGs. Expectation Maximization. Combining SCFGs with evolutionary models: the KnudsenHein algorithm. Lecture 11. Theory: Covarion Models as ‘profileSCFGs’. The RFam database. Predicting tRNAs in the human genome. Lecture 12. Theory: The ZukerTinoco model for RNA secondary structures. Calculating the partition function of the Boltzmann distribution and other moments of the Boltzmann distribution. COMPUTATIONAL NEUROSCIENCE
1. General introduction to Computational Neuroscience 2. General introduction to the anatomy, evolution and cellular basis of the nervous system 3. Basics of nerve cell electrochemistry and electrophysiology. Conductance based models of neurons 4. Parallel conductance model. Mechanism of action potential generation. The HodgkinHuxley model. Ionic currents, ionchannels, gate kinetics 5. Simplified neuron models. Simplifications of the HodgkinHuxley model: the FitzHughNagumoRinzel model, phasespace analysis. Explanation of bursting by bifurcation analysis. Abstract models: phase model, rate model, McCullochPitts neuron, integrate and Fire neuron model 6. Beyond the Hodgkin_Huxley model. Diverse voltage and ligand gated kinetics in singlecompartment models. Role of cellular morphology, dendritic effects. The cableequation and multi compartmental models. What is detailed modeling good for? Taxonomy of neuron models. Synapses and synaptic plasticity. Detailed, simplified and phenomenological models od the synaptic function 7. Cellular bases of learning: synaptic plasticity. The Hebbian rule of learning. variations for the Hebbian rule. Long term synaptic potentiation and depression. Synaptic plasticity on different time scales. Metaplasticity. Basics of modeling neural networks. The two (three) levels of neural dynamics. Learning rules: reinforcement, supervised and unsupervised learning. Basic neural architectures: feedforward and feedback structures, lateral connections, attractor networks 8. Windows to the World: traditional and modern measuring and data processing techniques: EEG, PET, fMRI, electrodes, intra and extracellular measurements, patchclamp. Fourier and wavelet transformations, EEG/MEG imaging, spikesorting 9. Neural oscillations: generation of oscillations an the cellular and network level. Oscillation based neural computations: timing and dynamic linking. Oscillations in memory models 10. The hyppocampus: modeling memory and spatial navigation. Place cells and place fields. Phase and rate coding. Dynamic modes of the hyppocampus 11. Modeling neurological and psychiatric disorders. Epilepsy, Parkinson's disease, Alzheimer's disease, schisophrenia. PROBABILISTIC MODELS OF THE BRAIN AND THE MIND
Machine learning, unsupervised learning, Bayesian networks, reinforcement learning, sampling algorithms, variational methods, computer vision, Cognitive science, inductive reasoning, statistical learning, semantic memory, vision as analysis by synthesis, sensorimotor control, classical and instrumental conditioning, behavioural economics, Neuroscience, neural representations of uncertainty, probabilistic neural networks, probabilistic population codes, natural scene statistics and efficient coding, neuroeconomics, neuromodulation COMPUTATIONAL NUMBER THEORY
Primes, primes of special form, the prime number theorem Sieving Arithmetic on large numbers Primality test: Fermat and Frobenius test Proving primality: the polynomial algorithm Factoring primes: Pollard rho method, Babystep Giantstep method Solving the discrete logarithm problem Subexponential pactoring algorithm, quadratic sieve Elliptic curves, using elliptic curves in factoring PROTOCOLS Prerequisites: Basic Algebra 1, Introduction to Computer Science
What protocols are; properties, attacks agains protocols Turing Machines, Oracle, computational and decisional DiffieHellman problems Protocols involving two or more parties, security notions, simulability ZK protocols, concurrent ZK, resettable ZK protocols, AM protocols Public key and symmetric key protocols Protocols for key establishment Identity based protocols, signatures Famous attacks against famous protocols 