New User Special Price Expires in

Let's log you in.

Sign in with Facebook


Don't have a StudySoup account? Create one here!


Create a StudySoup account

Be part of our community, it's free to join!

Sign up with Facebook


Create your account
By creating an account you agree to StudySoup's terms and conditions and privacy policy

Already have a StudySoup account? Login here


by: Mozell Lind


Mozell Lind
Texas A&M
GPA 3.58

Y. Choe

Almost Ready


These notes were just uploaded, and will be ready to view shortly.

Purchase these notes here, or revisit this page.

Either way, we'll remind you when they're ready :)

Preview These Notes for FREE

Get a free preview of these Notes, just enter your email below.

Unlock Preview
Unlock Preview

Preview these materials now for free

Why put in your email? Get access to more of this material and other relevant free materials for your school

View Preview

About this Document

Y. Choe
Class Notes
25 ?




Popular in Course

Popular in ComputerScienence

This 21 page Class Notes was uploaded by Mozell Lind on Wednesday October 21, 2015. The Class Notes belongs to CPSC 420 at Texas A&M University taught by Y. Choe in Fall. Since its upload, it has received 20 views. For similar materials see /class/226082/cpsc-420-texas-a-m-university in ComputerScienence at Texas A&M University.

Similar to CPSC 420 at Texas A&M




Report this Material


What is Karma?


Karma is the currency of StudySoup.

You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!

Date Created: 10/21/15
CPSC 420500 Artificial Intelligence Instructor Yoonsuck Choe choetamuedu httpcoursescstamueduchoe08fall420 Office HRBB 3228 TH 200pm 300pm TA Lei He Office TBA All communications out of the class will be through email registered on NEO neotamuedu and the announcements on the web page so regularly check out the web page Class notes will be available on the web 24 hours prior to the class It is your responsibility to print it out and bring it to the class 39 I lam Things You May Need Students with disability Please contact the department office H RBB 3rd floor for assistance See the syllabus forthe full information Computer UNIX accounts If you don t have one get one httpwwwcstamuedudepartment policiesaccounts Getting Your Money s Worth Utilize your instructor and TA as much as possible You have paid for their services Syllabus httpcoursescstamueduchoeO8fall420 0 See handout which is just a hardcopy of the web page on day 1 What is Intelligence Textbook Definitions 0 Thinking like humans 0 Acting like humans 0 Thinking rationally 0 Acting rationally lt However it depends on the definition whatever the intelligence test tests What is Al A folk popular view of Al From httpwww2cscmueduafscscmuedu userzhuxjwwwtravelfun imagesterminatorjpg 10p Universal studio s movie Terminator bottom Approaches Two basic stances 0 Strong Al 1 Build something that actually thinks intelligently 2 Simulation of thought would give rise to the pheonmenology of thought ie how it feels like to think 0 Weak Al 1 Build something that behaves intelligently 2 Not worried about its feelings But Really What is Al Diverse areas httpwwwaaaiorg Problem solving Reasoning Theorem proving Learning Planning Knowledge representation Perception and Robotics Agents and more 6 Problems Strong Al Hard to determine if something is really consciously intelligent or not the other minds problem in philosophy Weak Al Utility of the result is limited by the stated goal Hard to achieve a general usefulness as in true intelligence How to do Al Why not Follow the PlaneModel Why not engineer Al in the same way people engineered airplanes Certain things may seem physically impossible in terms of efficiency 139 Flight etc eg the flight of flies goal is simple 0 Flapping their wings cannot generate enough lift for their body 0 You know when a thing is flying weight but they do fly 2 Intelligence 0 Jet turbines cannot explain how the beetles achieve such an goal is complex and hard to define clearly impossible feat 0 Intelligence is a collection of many abilities I 0 Recent observation There are many ways to meet a single clear goal flight but there can Flies gyrate their wings to generate a vortex to create greater lift be only a small number of ways to simultaneously meet a huge number I I I I I I I I Moral if you fail to burld the impossible study an exrsting solution of unclearly defined goals intelligence How Flies Fly Then How to do Al Instructor s perspective Leadmgedge vortex V I 0 Importance of studying brain function 0 Induced 39 duwnwash Nona 0 Influence of environmental regularities on brain development and function 0 Interaction of the brain with the environment through action We must think about the more fundamental issues from time to time when research seems to be at a deadend Source httpwwwnaturecomnsuO1082301082310html Back to Reality Overview Lets be real39st39c39 i39 0 Related academic disciplines 0 Study strategies employed by humans in dealing with realworld 0 History of Al problems These include all the topics listed earlier 0 Hard Problems The background you learn in this course wrll enable you to Current Trends appreciate the deepness of the problems and to pursue further interest in Al and in human and machine intelligence in general 13 14 Foundations of Al Philosophy of Mind I The mindbody problem 0 Philosophy 0 Dualism Mind and body are separate entities 0 Mathematics 0 Monism Only mind or body exist but not both 39 PSVChOIOQV 1 Idealism all things are mental Cognitive Science 2 Materialism all things are material Linguistics 0 Epiphenomenalrsm mental phenomena are Just sideeffects of physical change in the brain ie they 0 Neuroscience do not have causal power over behavior Too many variations to mention all 15 16 Mathematics Algorithm alKhowarazmi Boole Hilbert Godel Incompleteness theorem Turing Halting problem Cook and Karp P NP and the like RepresentationInterpretation SymbolComputing the computersoftware metahpore Linguistics WW II machine translation Phonetics syntactic theory semantics discourse etc Innate vs learned Chomsky Syntax finite automata context free grammar etc Semantics semantic nets Subsymbolic selforganizing maps episodic memory recurrent neural nets etc Psychology Behaviorism stimulusresponse and conditioning Functionalism internal representations and processes Implementation independent Perceptual psychology vision audition etc Cognitive psychology cognition as information processing Holistic vs localist debate emergent vs simple summation Cognitive Science Interdisciplinary field studying human perception and cognition ranging over 0 Neuroscience 0 Behavioral science 0 Social science 0 Psychology 0 Computational science 0 Information theory 0 Cultural studies Ges Neuroscience Staining Golgi Nissl Hubel and Wiesel orderly structure of cat visual cortex PET scans and CAT scans localizing functional modules fMRl imaging cognitive and perceptual tasks Optical imaging orderly structure TMS zap and numb your brain History of Al I tation 1943 1956 McCulloch and Pitts early neural nets Minsky and Papert limitations of perceptron Newell and Simon physical symbol system hypothesis Logic Theorist Dartmouth Workshop 1956 Al was born It is 501 years old 2007 httpenwikipediaorgwikiAl50 23 Connections Scientific discoveries came from observing unexpected connections 0 Apple and gravity 0 Cloud chamber and the discovery of subatomic particles 0 Looms with punchcards and modern computers History of Al ll Early successes 1952 1969 0 General problem solver o McCarthy LISP 0 Toy domains ANALOGY STUDENT algebra o Widrow and Hoff adalines o Rosenblatt perceptrons History of Al Ill The 60 s and 70 s ELIZA Genetic algorithms Knowledgebased systems avoid the weak method ie search scientific domain engineering domain natural language The 80 s 5th generation Al Prolog 25 Hard Problems I Physicalism materialism and naturalism brain causes mind Functionlism if it functions in the same way a silicon brain can also be conscious Dualism and homunculus the Cartesian theatre Wide vs narrow content real correspondence or limited to experiential state lntentionality how can we believe in things that do not exist such as Poseidon History of Al IV 50th anniversary in 2006 httpenwikipediaorgwikiAl50 Some quotes from the 50th anniversary event Rodney Brooks the social sophisitication of 10yearold the manual dexterity of a 6yearold the language ability of 4yearold the visual object recognition of a 2 yearold Hard Problems ll Semantic content and syntactic symbols how can syntacic constructs posess intenionality Symbol grounding sensory devices produce grounded symbols and composite symbols can be constructed Problem of qualia why do we feel in such a way Turing test and Searle s Chinese Room system reply robot reply Hard Problems lll However the assumption that a collection of unconscious things are unconscious is invalid think about organic vs inorganic lite vsinanimate matter Searle s point of view mind is an emergent phenomena of the neural substrate biological naturalism Unix Environment Self Study The following slides are FYI they won t be covered in the class It you have any question about these please see the instructor or the TA Remote access Shell Files and Directories File and Dir Permissions Customizing the environment Execution of programs Getting More Information Editor pico Current Trends Learning instead of handcoding or strict reasoning Neural networks and statistical methods Genetic algorithms Evolutionary algorithms Embodied robotics Dynamical systems approach Bioinformatics Computational Neuroscience Distributed Agents Some thoughts on consciousness Crick and Koch 30 Remote Access Telnet insecure vs SSH secure httpwwwtreesshorg For windows use PUTTYEXE use the SSH mode Oncampus suncstamuedu interactivecstamuedu etc Offcampus Only by using TAMU VPN Use TAMU vpn to access other sunhosts Shell 0 Like DOSPROM PT but much more powerful 0 Several variations csh sh tcsh bash zsh ksh o tcsh derived from csh is your default opsFinding out your default shell sun gt ps PID TTY TIME CMD 964 ptsl2 003 tcsh File and Directory Cmds 0 Directory Listing ls short DlR W ls 7a list dot files also ls fl long DlR ls it newest file first ls 7F mark with dir sym links ie shortcut executable k etc 0 Any combination is allowed ls alF 0 Changing direcotry col dirname Files and Directories 0 Same as in MSDOS except longer filenames case sensitive 0 Path delimiter is not userchoefilename 0 Why MSDOS differs They use for command options FORMAT S The most important directory is your home directory useryouridhere Cont d 0 Creation and deletion mkdir dirname rmdir dirname Filecopy cp srcfile destfile cp srcfile destdir Moving and renaming mv srcflle destflle renamefile mv srcflle destdir move file to different directory File and Dir Perms unlxiNgt ls 71 tota1 568 1 shoe staff 22016 0st 1 1126 Adobantlst l Choe Staff 600 Nov 20 1737 PUTTYRND 2 shoe staff 4096 Jan 14 2224 RCS 1 shoe staff 4869 Jan 9 0934 aaaiOZ 2 shoe staff 4096 Jan 13 1545 asst d Twm mx Twm v 4 1 11 or file Owner group other r read w write x executeallow cdlordrr permission Eerironment Variables Envrronment varra es used to configure your env setenv gives you the list with their values setenv VARNAMEvalue to define or reset Append to get the value echo VARNAME to view value Shell variables use 3 et Home directory echo HOME This is where all the action begins Everything below here belongs to you almost Going to home col or Cd or Cd HOME or Changing File and Dir Perms chmod ulglo H7 rlwlx fileldir o chmod oirwx file 0 You can also use octal rwxrixrix 111101101 binary 755 octal o chmod 600 filename IWX iiii ii 0 chmod 755 dirname IWXI XI X Directory needs to be readable to do 1 s and executable to do col Customizing Your Shell n tcsh o login only executed when you first login 0 cshrc run everytime you create asubshell ie running a shell within a shell 0 Basically a list of shell commands like the BAT file Just add these lines in login setan PATH quotHOMEbinPATHquot set promptquotm gt quot alias ls quotls 7Fquot Execution of Programs Getting More Info 0 Use the man command 0 ls is actually usrbinls man ls 0 Because PATH contains usrbin the above file gets man chmod executed man mkdir o For local executables compiled ones etc in the current dir add 39 NaVigating in man Pages to specifythe current dir execfilename space or CTRL F pgdown o This is not a concern because well be using GCL interpreter CTRLiB ngp usrlocalbingcl ltStringgtENTER3 search fwd for ltstringgt ltstringgtENTER search bwd for ltstringgt or alonerepeatsearch fwd orbwd 41 42 Editor PICO Tosh tips Opt CSWbil l piCO 0 Command and filename completion just type a partial string and An easy to use text editor navigate with arrow keys and press tab CTRL7Ypgup and CTRLilpgdn CTRL X Exit answer carefullyl CTRL O Save As CTRL R Read another file into current position 0 Use arrow keys to go backward like in doskey CTRL K Cut you can cut multiple lines CTRL U Paste one or more lines cut with CTRLK CTRL W Search Where is No undo I recommend vi or emacs Of course you may ftp Pipes My output is your input output of ls goes to grep sunuserchoegt ls ial l grep no 3 choe staff 4096 Oct 10 1053 drwx 7777 if ssh27no irw 77777 77 l choe staff 123 Jan 14 1332 moms drwx 7777 if 2 choe staff 4096 Nov 12 1441 notes Little Bit of LISP httpwwwcstamuedutacultychoe coursesO5talllispquickrethtm o CMUCL Carnegie Mellon University Common LISP o optappscmuclbir1lisp 0 At the prompt just type the expressions unix Ngt lisp CMU Common Lisp cvs Head 2003707701 162301 running on unix usrlocalllbcmuclllbllsp core With Core 200307p01 1601OO05OO on emplcS Dumped on Tue See lthttp www cons orgcmuclgt for support information Loaded subsystems Python 11 target UltraSparcSolaris 7 CLOS based on Gerd s PCL 20030618 092309 102m x i quit Ngt unlx 47 Useful Commands o grep pattern matching also awk and sed grep ltpatterngt ltfilenamegt ltsomeiothericommandgt I grep ltpatterngt 0 find find file with a certain pattern find ltstart7dirgt ltpatterngt sprint 0 we line word and char count ls ial we 0 df disk partitions and current usage 0 du disk usage under a directory Other calzcalendar datezdate uname 7aOSversion etc Next Time 0 Lisp CPSC 420 500 Program 3 Perceptron and Backpropagation Yoonsuck Choe Department of Computer Science Texas AampM University October 31 2008 1 Overview You Will implement perceptron learning from scratch see section 3 for details and train it on AND OR and XOR functions Then you Will take an existing backprop agation code see section 4 train it and test it under different conditions and report your ndings The same three boolean functions Will be used for the backpropagation learning For further details on perceptrons and backpropagation see the lecture slides and also Hertz et al 1991 Speci c submission instruction Will be given in section 5 and section 6 2 Language and OS You may use either CC Java Matlab or Octave or Lisp The resulting code should be able to compile and run on the departmental unix host unixcstamuedu You may use a different language With a permission from the ins uctor in Which case you Will be asked to do a demo in front of the TA To compile your programs other than Lisp Which you already know see the following instructions 0 C program le perceptronc to compile cc 0 perceptron perceptronc 1m to execute lperceptron 0 C program le perceptronC to compile c 0 perceptron perceptronC 1m to execute lperceptron 0 Java program le perceptronjava to compile javac perceptronjava to execute java perceptron The full paths for the compilers are on computecstamuedu o usrbincc o optcswgcc3bing o usrjdklatestbinj ava o usrjdklatestbinj avac 3 Perceptron Perceptron activation is de ned as i0 2 Output step INPM l where stepX l ifX 2 0 and stepX 0 ifX lt 0 see gure 1 XXX ww m W2 1 X Y Figure l Perceptron The input I N P0 is the bias unit xed to 71 and has an associated weight WO which is the threshold The two inputs X and Y are given and the output fX Y will be calculating or attempting to calculate a boolean function one of OR AND or XOR Implement a perceptron with two input units three including the bias unit that has a xed input value 1 and one output unit Your program should take three inputs from the command line 1 maximum number of epochs to run integer 2 learning rate parameter a value double 3 function selection string and or and xor For example a typical run would go like this S is the unix prompt S perceptron 10000 00001 and A pseudo code for perceptron learning is as follows 1 Initialize weights to random numbers between 00 and 10 Note that you have just one set of 3 weights W0 Wl W2 You will train this same set using the four inputtarget pairs 2 Initialize epoch count to 0 3 while sum of target 7 output2 for all input pattenis is not 0 do 0 for each inputtarget patteni a present input and calculate output b calculate the error target 7 output c update the weights using the perceptron learning rule 0 endfor o increment epoch count 0 if epoch count gt max epochs break from while loop 4 Print out the output for the inputs 00 01 10 and 11 4 Backpropagation For backpropagation you will download the following le make it one line http faculty cs tamu educhoe srcbackpropil6targz and run it under different conditions First you need to unzip and untar it by running the following S tar xzvf backpropil6 tar gz Cd backprop then read the README le to learn how to compile and run it Running the bp program which is generated by compiling will give you a huge dump on the screen To selectively View the data you re more interested in use the grep command For example to View the progression of error S bp confXorconf l grep ERR and to View the actual output values for the inputs S bp confXorconf l grep OUT 5 Assignment This section will detail what you actually have to do and have to submit All projects should be tunied in using the c snet tu rm 1 n 51 Perceptron Implement perceptron leaniing algorithm as detailed in section 3 and with the program conduct the following experiments and submit the required material along with the code and the README le as usual Experiments 1 Test AND OR and XOR for learning rates a 0001 and 00001 For each Boolean function run the experiment with different initial random weights use the random number generator function to do this two times Discard all runs that ended in 1 epoch you will see several of these why would they occur The total number of trial will thus be 2 leaniing rates X 2 random initial weights X 3 functions to leani 12 Set the max epoch to 10000 for all trials 2 For each trial report the following in the README le a initial weights and show the plot of the decision boundary the straight line de ned by the weights b nal weights and show the plot of the decision boundary c number of epochs taken to complete training if successful let us call this n and d for each epoch the sum of squared error for all four input pattenis 3 Answer these questions in the README le a Do you think perceptron will be able to learn the boolean function f X Y X V Y What do you think is the role of the Sign of the weights in this case Think about the geometric interpretation in that case Table 1 Boolean Function fXY X V Y X Y X VY 52 Backpropagation With the provided code conduct experiments on AND OR and XOR Note that in the bp cc code learning rate a is a named eta just in case you want to take a look inside the code Experiments 6 1 Test AND OR and XOR for learning rates a 001 and 0001 and plot the sum of squared error for each trial a total of 6 trials A total of 3 set of plots for AND OR and XOR with each set containing 2 curves for two a s is re quired Save the plots in image les and name them andl jpg and2 jpg 011 jpg 012 jpg xorl jpg X0r1jp Use grep to ex tract the error values see below and load that le dump txt in a spread sheet S bp confXorconf 1 grep ERR gt dumptXt Q With learning rate a 001 test AND OR and XOR with 1 2 and 4 hid den units Plot the sum of squared error for each trial A total of 3 set of plots for AND OR and XOR with each set containing 3 curves for three different number of hidden units is required S ave the plots in image les and name them bpiandl jpg bpiandZ jpg bpiand3 jpg L For all trials above count the number of epochs until the end is reached and measure the time taken using the timex unix command add user time and sys time timeX bp confXor conf Report the number of epochs and time spent for each trial in the README le How many number of hidden units was the best in your opinion and why Submission Details The due date is 1242008 1159pm Grading criteria will be similar to the previous programming assignments You must submit the following 0 source code 0 compiled executable binary o README le containing material detailed in section 5 and 0 plots image les include a list of image les and a brief description of each in the README le 0 put everything in a single zip or targz le Only electronic submission through CSNET Will be accepted References Hertz Krogh A and Palmer R G 1991 Introduction to the Theory ofNeural Computation Reading MA AddisonWesley CPSC 420 500 Paper Commentary Instructions Due 12307 in class Yoonsuck Choe November 14 2007 1 Goals The main purpose of this assignment is to expose you to recent literature on Al and related elds Another major goal is to help you build the ability to actively understand analyze evaluate and critique other people s ideas not just passively absorb knowledge Finally this assignment will help you organize your thought and express your ideas in a coherent manner through writing 2 Selection of a Paper Read the instructions in http Courses cs tamu educhoe42 Oireading if you want to select a paper on your own Otherwise select one from the following H Fundamental issues Bell 1999 N Behaviorbased vs information processing view of intelligence Dean 1998 9 Symbol grounding and natural semantics Cohen and Beal 2000 Choe and Smith 2006 Dynamical systems approach in cognitive science Beer 2000 Cognition and selforganization Langlois and Garrouste 1997 0915 If you want to read something else please obtain permission from the instructor 3 Content of the Review The paper review must address the factual content of the paper and your interpretation and critique More speci cally it should contain all of the following aspects 1 A brief summary of the paper What is the main lessonmessageconclusion of the paper 1 para graph 2 Main contribution of the paper What is new about the approach taken in this paper ie why is it distinct from other work andor what are the new issues it raises compared to other cited related work andor what is new about the view presented in this paper 1 paragraph 1 3 Limitations and future directions What are the limitations of the approach and what kind of issues need to be addressed in the future 1 paragraph 4 Relevance How is the work different from or similar to the AI methods we studied throughout this semester 1 paragraph All parts of the review should be in your own words You should not use verbatim copy of the abstract or conclusion or any other part of the paper The review should not exceed one typed singlespaced page 4 Grading Criteria The submitted reviews will be graded according to the following criteria 1 Clarity Is the review clearly written 30 2 Succinctness Is the review brief and to the point ie no redundancy 10 It should not contain unnecessary or redundant passages 9 Accuracy Does the review accurately represent the factual content of the paper You should make sure that you understand the paper well and when you say something about the paper it should be factually correct If the paper uses too many technical terms and you need help please let me know 25 5 Depth and originality of the analysis Is the interpretation and analysis presented in the review insightful Is the argumentation sound 20 U Organization Is the review well organized This does not mean pretty formatting using word pro cessors Organization is how the chunk of ideas are ordered and structured 10 6 Remaining 5 see below Submission section Note that grammatical errors will nut in uence the grade unless they are severe This is not an English writing class However typographical errors will because it shows the lack of your attention to detail Also be aware that your opinion can differ from mine and that s ne as long as you present a reasonable argument Simply reiterating what the author s view is does not automatically guarantee you a good grade 5 Submission The paper review is due by 120307 in class For the paper review you need to submit two things 1 Printout of an early draft with your editorial corrections on it written on the printout You should carefully read the whole review at least once and revise it at least once based on your com ments After you revise read it again to make sure you didn t introduce any further error Your initial version with your own comments written on it will serve as an evidence that you did revised it at least once This will account for 5 of the grade Note Do not use MS Word s change tracking function as a substitute for this requirement 2 2 Printout of the nal version Note that electronic submission will not be accepted 6 Reviewing Tips If you are not sure what is a good review take a look at this book review by Cosma Shalizi http CSCS umich edu Ncrshalizireyiewshowi theimindiworks 7 Extra Credit Information You may submit one extra review for 4 extra credit toward your nal grade That makes it a maximum of two reviews References Beer R D 2000 Dynamical approaches to cognitive science Trends in Cagnitive Sciences 491799 Bell A J 1999 Levels and loops The future of arti cial and Iquot quot 39 39 Transactians afthe Rayal Saciety afLandan 354201372020 Choe Y and Smith N H 2006 Motionbased autonomous grounding Inferring external world prop erties from internal sensory states alone In Gil Y and Mooney R editors Praceedings 0f the 21st Natianal Canference an Arti cial Intelligence 9367941 Cohen P R and Beal C R 2000 Natural semantics for a mobile robot Technical Report 0059 University of Massachusettes Department of Computer Science Dean J 1998 Animats and what they can tell us Trends in Cagnitive Sciences 260767 Langlois R and Garrouste R 1997 Cognition redundancy and learning in organizations Ecanamics aflnnavatian andNew Technalagy 42877299


Buy Material

Are you sure you want to buy this material for

25 Karma

Buy Material

BOOM! Enjoy Your Free Notes!

We've added these Notes to your profile, click here to view them now.


You're already Subscribed!

Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'

Why people love StudySoup

Steve Martinelli UC Los Angeles

"There's no way I would have passed my Organic Chemistry class this semester without the notes and study guides I got from StudySoup."

Jennifer McGill UCSF Med School

"Selling my MCAT study guides and notes has been a great source of side revenue while I'm in school. Some months I'm making over $500! Plus, it makes me happy knowing that I'm helping future med students with their MCAT."

Jim McGreen Ohio University

"Knowing I can count on the Elite Notetaker in my class allows me to focus on what the professor is saying instead of just scribbling notes the whole time and falling behind."


"Their 'Elite Notetakers' are making over $1,200/month in sales by creating high quality content that helps their classmates in a time of need."

Become an Elite Notetaker and start selling your notes online!

Refund Policy


All subscriptions to StudySoup are paid in full at the time of subscribing. To change your credit card information or to cancel your subscription, go to "Edit Settings". All credit card information will be available there. If you should decide to cancel your subscription, it will continue to be valid until the next payment period, as all payments for the current period were made in advance. For special circumstances, please email


StudySoup has more than 1 million course-specific study resources to help students study smarter. If you’re having trouble finding what you’re looking for, our customer support team can help you find what you need! Feel free to contact them here:

Recurring Subscriptions: If you have canceled your recurring subscription on the day of renewal and have not downloaded any documents, you may request a refund by submitting an email to

Satisfaction Guarantee: If you’re not satisfied with your subscription, you can contact us for further help. Contact must be made within 3 business days of your subscription purchase and your refund request will be subject for review.

Please Note: Refunds can never be provided more than 30 days after the initial purchase date regardless of your activity on the site.