06 2019 档案
摘要：Linearly Separable Data Multiple Linear Questions Constructing a Decision Tree First Split Coding A Decision Tree Decision Tree Parameters Data Impuri
阅读全文
摘要：Robot Tour Guide Examples MDP Grid World Problems With Conventional Planning For this problem (and only this problem) assume actions are stochastic in
阅读全文
摘要：(PP 1.1) Measure theory: Why measure theory - The Banach-Tarski Paradox ball -> tear down to pieces -> rigid transformation (translation/rotation) ->
阅读全文
摘要：represent => reason => represent =>.... data => knowledge => data > ...... Current Research
阅读全文
摘要：Mistakes in Reasoning and Learning Beyond Mistakes: Knowledge Gaps The Blurred Line Between Cognition and Metacognition Strategy Selection Strategy In
阅读全文
摘要：Optional Readings : Questions for Correcting Mistakes Visualizing Error Detection Error Detection Algorithm Explanation-Free Repair Explaining the Mis
阅读全文
摘要：Optional reading: Stefik, Chapter 9 diagnose as classification and abduction Exercise: Diagnosing Illness Defining Diagnosis Problems with Diagnosis a
阅读全文
摘要：Exercise: Designing a Basement Defining Configuration The Configuration Process left top: task right bottom: solution task v1 Configuration Space > so
阅读全文
摘要：Optional Readings : Exercise: Recognizing 3D Figures Exercise: Gibberish Sentences Constraint Propagation Defined From Pixels to 3D Line Labeling Cons
阅读全文
摘要：Optional reading: Winston Chapter 20 Incremental Concept Learning Revisited Abstract Version Spaces Visualizing Version Spaces converge model is thre
阅读全文
摘要：Need for Cross-Domain Analogy should apply the laser light? kill the tumor and hurt the healthy tissue as well? decompose weaker lights Spectrum of Si
阅读全文
摘要：Optional Readings : Exercise: Transporting Soup for robot, it's hard to decide. Example: Retrieving a Cup Concept Space Prior Knowledge Abstraction Tr
阅读全文
摘要：This lesson borrows heavily from the following book: Rich, E., & Knight, K. (1991). Artificial intelligence. McGraw-Hill, New York. Optional reading:
阅读全文
摘要：how to prevent robot from jumping off through windows Example: Ashok Ate a Frog how to make machine understand they have the same meaning Primitive Ac
阅读全文
摘要：Optional Readings : The Earthquake Report Thematic Role Systems Constraints Resolving Ambiguity in Prepositions Ambiguity in Verbs Optional Reading: W
阅读全文
摘要：Block Problem Revisited Painting a Ceiling States Optional Reading: Winston Chapter 15, pages 323-336 Operators Planning and State Spaces plan: Planni
阅读全文
摘要：Why do we need formal logic? Inferences About Birds Winston Chapter 13: Page 284 Exercise: Inferences About Foos Predicates Conjunctions and Disjuncti
阅读全文
摘要：Optional Readings : Classifying Birds The Challenge of Classification combinatorial explosion! how to make reasonable decision in real time? Equivalen
阅读全文
摘要：Exercise: Identifying a Foo \ This example is adapted from the 'arch' example in the the following book: Winston, P. (1993). Artificial Intelligence (
阅读全文
摘要：how to use memory and reason Assumptions of Case-Based Reasoning Case Adaptation (1) Model of the World (2) Recursive reasoning Further reading: Vatta
阅读全文
摘要：Optional Readings : Learning by Recording Cases Case Retrieval by Nearest Neighbor Nearest Neighbor in k-Dimensional Space learning and memory(recordi
阅读全文
摘要：Optional reading: Winston Chapter 9, pages 179-182, 202-206 compare frame with sematic network and production system How do we make sense of a sentenc
阅读全文
摘要：This example is adapted from the following paper: Lehman, J. F., Laird, J. E., & Rosenbloom, P. S. (1996). A gentle introduction to Soar, an architect
阅读全文
摘要：Optional Reading : Exercise: The Block Problem This example is adapted from the following book: Winston, P. (1993). Artificial Intelligence (3rd ed.).
阅读全文
摘要：Optional reading: Winston Chapter 3, pp. 47-50 knowledge representation, problem solving (going to be discussed) , architectures Dumb Generators, Dumb
阅读全文
摘要：Knowledge Representations eg. F=ma Introduction to Semantic Networks The example above is adapted from Raven's test of progressive matrices. For furth
阅读全文
摘要：Optional Readings : Introduction to Computational Psychometrics Putting Online Learning and Learning Sciences Together: https://www.youtube.com/watch?
阅读全文
摘要：Characteristics of AI Problems Characteristics of AI Agents What is Knowledge-Based AI? Optional Reading: "What is AI anyway?" - Ashok K. Goel, Jim Da
阅读全文
摘要：Introduction: What is software engineering? Why do we need it? Importance of Software Engineering Evidence of the Software Crisis You can access the p
阅读全文
摘要：The RL problem Trading as an RL problem Mapping trading to RL Markov decision problems Unknown transitions and rewards What to optimize? Learning Proc
阅读全文
摘要：A closer look at KNN solutions What happens as K varies What happens as D varies Metric 1 RMS Error In Sample vs out of sample Which is worse? Cross v
阅读全文
摘要：a data-centric way to build predictive models The ML problem Supervised regression learning Robot navigation example How it works with stock data Exam
阅读全文
摘要：this lesson => Buffet said two things => (1) investor skill => (2) breadth / the number of investments Grinold's Fundamental Law breadth => more oppor
阅读全文
摘要：How data is aggregated tick interval can be day, mintue, ... etc. Price anomaly Stock splits Split adjustment Dividends Adjusting for dividends the cl
阅读全文
摘要：Risks for hedge funds Two stock scenario Two stock CAPM math how to remove the incluence of the market or beta??? Allocations remove market risk How d
阅读全文
摘要：What is a company worth? Yes, the value of $1 would typically reduce over time, and the sum of the individual $1's generated every year would converge
阅读全文
摘要：Computational Investing MC2: Computational Investing Lessons: In this mini-course, we focus on modeling the behavior of stock markets. Projects: Types
阅读全文
摘要：What is an optimizer? Minimization example How to defeat a minimizer Convex problems Building a parameterized model Minimizer finds coefficients What
阅读全文
摘要：A closer look at daily returns Histogram of daily returns gaussian => kurtosis = 0 How to plot a histogram Computing histogram statistics Select the o
阅读全文
摘要：Lesson outline Pandas makes it very convenient to compute various statistics on a dataframe: Global statistics: mean, median, std, sum, etc. [more] Ro
阅读全文
摘要：Lesson outline Lesson outline Here's an overview of what you'll learn to do in this lesson. Documentation links are for reference. Read in multiple st
阅读全文
摘要：Python vs R and Matlab You will be using the Python data analysis library pandas extensively. You will also occasionally need to dabble in NumPy, SciP
阅读全文
摘要：The mobility provided by wireless networks has made a huge difference in the lives of most citizens of the internet. However, several challenges have
阅读全文
摘要：Finally! This lecture covers the glue that holds the entire internet together. After some discussion of the theoretical possibilities available in a n
阅读全文
摘要：TCP is an interesting protocol for many reasons, one of which I find fascinating is that it can employ many different behavioral algorithms at the end
阅读全文
摘要：The algorithm behind TCP must deal with more than merely moving lots of bytes from application to application. It must also ensure that it acts in a m
阅读全文
摘要：Understanding the Transport Control Protocol (TCP) is essential to a thorough understanding of network communication and for any development of distri
阅读全文
摘要：As discussed in last lecture, UDP does not provide any reliability guarantees beyond being able to detect bit-level errors. Transport-layer protocols
阅读全文
摘要：This is the first lecture in a series exploring the Transport layer. This layer is responsible for logical communication between applications running
阅读全文
摘要：https://blog.csdn.net/reyleon/article/details/12976889 https://www.sysgeek.cn/linux-dig/ https://www.thegeekstuff.com/2012/02/dig-command-examples/ 10
阅读全文
摘要：In order to analyze and debug networks, we often have to use tools that are mathematical in nature. This lecture will give you a basic understanding o
阅读全文
摘要：The growth of peer to peer technologies has had a profound effect on the internet over the past decade. These systems have pioneered new media distrib
阅读全文
摘要：The Domain Name System has become the largest distributed database in the world. Human usage of the internet would be difficult to imagine without som
阅读全文
摘要：We begin our examination of the application layer with one of the most ubiquitous application protocols: the Hypertext Transport Protocol. We will tea
阅读全文
摘要：http://www.ini740.rocks/F18/assignments/Lab0.pdf
阅读全文
摘要：Interactions between different level peers When trying to understand the behavior of the internet, one cannot ignore the business relationships of the
阅读全文
摘要：Internetworking with Gateways We will turn the philosophical and theoretical discussion from last lecture into a practical examination of the TCP/IP a
阅读全文
摘要：Sources of network delay This lecture describes many of the design decisions that have to be made to generate an architecture for inter-networking. Th
阅读全文
摘要：http://www.ini740.rocks/F18/lectures/class_1.html Lecture 1: Networking Introduction This lecture will introduce the course, with information about lo
阅读全文
摘要：Sample Final Solutions Page 1 of 8 CS 8803, Spring 2015. 1. Timeslices Problem On a single CPU system, consider the following workload and conditions:
阅读全文
摘要：Quiz Note We got our estimates from here. Current estimates have continued to see growth. Internet Services Internet Service Architectures “Homogeneou
阅读全文
浙公网安备 33010602011771号