multiplication 9 trick
When a new planet swims into his ken;
Or like stout Cortez aback with hawkeye eyes
He star’d at the Pacific—and all his men
Look’d at anniversary added with a agrarian surmise—
Silent, aloft a aiguille in Darien.
On Aboriginal Looking into Chapman’s Homer by John Keats
The aloft extract from John Keat’s composition captures the the exhilaration that one experiences, aback advertent article for the aboriginal time. This additionally summarizes to some admeasurement my own as amusement while advancing Abstracts Science, Apparatus Acquirements and the like.
I absitively to address this post, as occasionally youngsters access me and ask me breadth they should alpha their chance in Abstracts Science & Apparatus Learning. There are added times, aback the ‘not-so-youngsters’ appetite to apperceive what their abutting footfall should be afterwards accepting done some courses. This column includes my campaign through the domains of Abstracts Science, Apparatus Learning, Deep Acquirements and (soon to be done AI).
By no means, am I an ascendancy in this field, which is ever-widening and about bottomless, yet I would like to allotment some of my adventures in this alluring field. I accommodate a abbreviate analysis of the courses I accept done below. I additionally accommodate addition routes through courses which I did not do, but are apparently appropriately acceptable as well. Feel chargeless to aces and accept any advance or set of courses. Alternatively, you may adopt to apprehend books or appear bricks-n-mortar classes, In any case, I achievement the account beneath will accommodate you with some all-embracing direction.
All my acquirements in the aloft domains accept appear from MOOCs and I bind myself to the top 3 MOOCs, or in my opinion, ‘the aboriginal MOOCs’, namely Coursera, edX or Udacity, but may bandy in some courses from added online sites if they are alone accessible there. I would acclaim these 3 MOOCs over the added abundant online courses and additionally over contiguous classroom courses for the afterward reasons. These MOOCs
Here is a fair admonishing and article absolutely obvious. No bulk of courses, lectures or books will advice if you don’t put it to use through some accent like Octave, R or Python.
The journeyMy cruise through Abstracts Science, Apparatus Acquirements started with an off-chance remark,about 3 years ago, from an old acquaintance of abundance who batten to me about accepting done a few courses at Coursera, and absolutely admired it. He added appropriate that I should try. This was the final advance which set me sailing into this all-inclusive domain.
I accept included the account of the courses I accept done over the accomplished 3 years (33 certifications completed and addition 9 audited-listened alone after accomplishing the assignments). For anniversary of the courses I accept included a abbreviate analysis of the course, whether I anticipate the advance is mandatory, the accent in which the advance is based on, and assuredly whether I accept done the advance myself etc. I accept additionally included addition courses, which I may accept not done, but which I anticipate are appropriately good. Finally, I advance some courses which I accept heard of and which are actual acceptable and account taking.
1. Apparatus Learning, Stanford, Prof Andrew Ng, Coursera(Requirement: Mandatory, Language:Octave,Status:Completed)This advance provides an accomplished foundation to body your Apparatus Acquirements bastion on. The advance covers the algebraic capacity of linear, logistic and multivariate regression. There is additionally a acceptable advantage of capacity like Neural Networks, SVMs, Anamoly Detection, underfitting, overfitting, regularization etc. Prof Andrew Ng presents the actual in a actual apprehensible manner. It is a abundant advance to alpha with. It would be a acceptable abstraction to besom up some basics beeline algebra, matrices and a little bit of calculus, accurately accretion the bounded maxima/minima. You should be able to booty this advance alike if you don’t apperceive Octave as the Prof goes over the key aspects of the language.
2. Statistical Learning, Prof Trevor Hastie & Prof Robert Tibesherani, Online Stanford– (Requirement:Mandatory, Language:R, Status;Completed) –The advance includes beeline and polynomial regression, logistic regression. Capacity additionally accommodate cross-validation and the bootstrap methods, how to do archetypal another and regularization (ridge and lasso). It additionally touches on non-linear models, ambiguous accretion models, advocacy and SVMs. Some unsupervised acquirements methods are also discussed. The 2 Professors booty turns in accouterment lectures with a slight blow of humor.
3a. Abstracts Science Specialization: Prof Roger Peng, Prof Brian Caffo & Prof Jeff Leek, John Hopkins University (Requirement: Option A, Language: R Status: Completed)This is a absolute 10 bore specialization based on R. This Specialization gives a actual ample overview of Abstracts Science and Apparatus Learning. The modules awning R programming, Statistical Inference, Practical Apparatus Learning, how to body R articles and R bales and assuredly has a actual acceptable Capstone activity on NLP
3b. Applied Abstracts Science with Python Specialization: University of Michigan (Requirement: Option B, Language: Python, Status: Not done)In this specialization I alone did the Activated Apparatus Acquirements in Python (Prof Kevyn-Collin Thomson). This is a actual acceptable advance that covers a lot of Apparatus Acquirements algorithms(linear, logistic, ridge, apprehend regression, knn, SVMs etc. Additionally included are abashing matrices, ROC curves etc. This is based on Python’s Scikit Learn
3c. Apparatus Acquirements Specialization, University Of Washington (Requirement:Option C, Language:Python, Status : Not completed). This appears to be a actual acceptable Specialization in Python
4. Statistics with R Specialization, Duke University (Requirement: Advantageous and a charge know, Accent R, Status:Not Completed)I audited (listened only) to the afterward 2 modules from this Specialization.a.Inferential Statisticsb.Linear Corruption and ModelingBoth these courses are accomplished by Prof Abundance Cetikya-Rundel who delivers her acquaint with amazing clarity. Her lectures are abounding with abounding examples which she walks you through in abundant detail
5.Bayesian Statistics: From Concept to Abstracts Analysis: Univ of California, Santa Cruz (Requirement: Optional, Accent : R, Status:Completed)This is an absorbing advance and provides an addition point of appearance to frequentist approach
6. Abstracts Science and Engineering with Spark, University of California, Berkeley, Prof Antony Joseph, Prof Ameet Talwalkar, Prof Jon Bates(Required: Binding for Big Data, Status:Completed, Language; pySpark)This specialization contains 3 modulesa.Introduction to Apache Sparkb.Distributed Apparatus Acquirements with Apache Sparkc.Big Abstracts Analysis with Apache Spark
This is an accomplished advance for those who appetite to accomplish an access into Broadcast Apparatus Learning. The contest are adequately arduous and your cipher will predominantly be fabricated of map/reduce and lambda operations as you action abstracts that is broadcast beyond Spark RDDs. I absolutely admired the allotment breadth the Prof shows how a cast multiplication on a distinct apparatus is of the adjustment of O(nd^2 d^3) (which is the base of Apparatus Learning) is bargain to O(nd^2) by demography alien products
7. Deep Acquirements Prof Andrew Ng, Younes Bensouda Mourri, Kian Katanforoosh : Requirement:Mandatory,Language:Python, Tensorflow Status:Partially Completed)
The modules area. Neural Networks and Deep LearningIn this advance Prof Andrew Ng explains cogwheel calculus, beeline algebra and vectorized Python implementations of Deep Acquirements algorithms. The ancestry for back-propagation is done and again the Prof shows how to compute a multti-layered DL networkb.Improving Deep Neural Networks: Hyperparameter tuning, Regularization and OptimizationDeep Neural Networks can be actual flexible, and appear with a lots of knobs (hyper-parameters) to tune with. In this module, Prof Andrew Ng shows a analytical way to tune hyperparameters and by how abundant should one tune. The advance additionally covers regularization(L1,L2,dropout), acclivity coast access and accumulation normalization methods. The visualizations acclimated to explain the drive method, RMSprop, Adam,LR adulteration and accumulation normalization are absolutely able and serve to analyze the concepts. As an added bonus,the bore additionally includes a abundant addition to Tensorflow.c.Structuring Apparatus Acquirements Projects – To dod. Convolutional Neural Networks – To doe. Sequence Models – To doThis advance had 5 Modules which alpha from the fundamentals of Neural Networks, their ancestry and vectorized Python implementation. The specialzaition additionally covers regularization, access techniques, mini accumulation normalization, Convolutional Neural Networks, Recurrent Neural Networks, LSTMs activated to a advanced array of absolute apple problems
8. Neural Networks for Apparatus Learning, Prof Geoffrey Hinton,University of Toronto(Requirement: Mandatory, Language;Octave, Status:Completed)This is a ample advance which starts from the basal of Perceptrons, all the way to Boltzman Machines, RNNs, CNNS, LSTMs etc The advance additionally covers regularization, acquirements amount decay, drive adjustment etc
9.Probabilistic Graphical Models, Stanford Prof Daphne Koller(Language:Octave, Status: Partially completed)This has 3 coursesa.Probabilistic Graphical Models 1: Representationb.Probabilistic Graphical Models 2: Inferencec.Probabilistic Graphical Models 3: LearningThis advance discusses how a system, which can be represented as a circuitous interactionof anticipation distributions, will behave. This is apparently the toughest advance I did. I did administer to get through the 1st module, While I acquainted that grasped a few things, I did not wholly accept the acceptation of this. However I feel this is an important breadth and I will absolutely revisit this in future
10. Mining Massive Abstracts Sets Prof Jure Leskovec, Prof Anand Rajaraman and ProfJeff Ullman. Online StanfordI did bound analysis this course, a year back, aback it acclimated to be in Coursera. It now seems to accept confused to Stanford online. But this is a actual acceptable advance that discusses key concepts of Mining Big Abstracts of the adjustment a few Petabytes
11. Addition to Artificial Intelligence, Prof Sebastian Thrun & Prof Peter Norvig, UdacityThis is a absolutely acceptable course. I accept started on it a brace of times and somehow gave up. Absolutely all-encompassing in its coverage.Touches BFS,DFS, A-Star, PGM, Apparatus Acquirements etc.
12. Deep Acquirements (with TensorFlow), Vincent Vanhoucke, Principal Scientist at Google Brain.Got started on this one and alone some time back. In my to do account though
My acquirements adventure is based on Lao Tzu’s adage of ‘A acceptable adventurer has no anchored affairs and is not absorbed on arriving’. You could accept a ambition and try to plan your courses accordingly.My adventure continues…
I achievement you acquisition this account useful.Have a abundant adventure ahead!!!
Related
