ML Math Skills - Data Science Central2020-07-10T03:54:20Zhttps://www.datasciencecentral.com/forum/topics/ml-math-skills?commentId=6448529%3AComment%3A722462&feed=yes&xn_auth=noI am in the same boat. Do I s…tag:www.datasciencecentral.com,2018-05-18:6448529:Comment:7226572018-05-18T05:56:27.675ZPradeep Sundaramhttps://www.datasciencecentral.com/profile/PradeepSundaram
<p>I am in the same boat. Do I started linear algebra from MIT Open Courseware's Linear algebra course by Prof Strang.</p>
<p>I am in the same boat. Do I started linear algebra from MIT Open Courseware's Linear algebra course by Prof Strang.</p> thanks Anil Kumar said:
Here…tag:www.datasciencecentral.com,2018-05-17:6448529:Comment:7224622018-05-17T22:07:05.887ZRavi Krishnappahttps://www.datasciencecentral.com/profile/RaviKrishnappa
<p>thanks<br></br> <br></br> <cite>Anil Kumar said:</cite></p>
<blockquote cite="https://www.datasciencecentral.com/forum/topics/ml-math-skills#6448529Comment720906"><div><div class="xg_user_generated"><p>Here is what I recommend when I'm mentoring and had good success with it<br></br><br></br>Ian GoodFellow is fairly advanced.<br></br><br></br>I would suggest you start with Grokking Deep Learning by Andrew Trask.<br></br>This book has the lowest barrier to entry possible in author's own words and I can attest he has…</p>
</div>
</div>
</blockquote>
<p>thanks<br/> <br/> <cite>Anil Kumar said:</cite></p>
<blockquote cite="https://www.datasciencecentral.com/forum/topics/ml-math-skills#6448529Comment720906"><div><div class="xg_user_generated"><p>Here is what I recommend when I'm mentoring and had good success with it<br/><br/>Ian GoodFellow is fairly advanced.<br/><br/>I would suggest you start with Grokking Deep Learning by Andrew Trask.<br/>This book has the lowest barrier to entry possible in author's own words and I can attest he has achieved his mission.<br/><a rel="nofollow noopener" href="https://www.manning.com/books/grokking-deep-learning" target="_blank">https://www.manning.com/books/grokking-deep-learning</a><br/>He breaks down all the concepts into extremely small and simple Intuition.<br/>you will be doing Gradient Descent & Back Propagation in your sleep :)<br/><br/>After that, I would read <span>Deep Learning with Python by François Chollet, creator of Keras<br/></span><a rel="nofollow noopener" href="https://www.manning.com/books/deep-learning-with-python" target="_blank">https://www.manning.com/books/deep-learning-with-python</a><br/>Now you will be armed with DNN, ConvNet, RNN, GAN, and ML(Linear Regression, Logistic & Kernel Methods). <br/><br/>For Math, here are some resources I'm listing in relevance to Deep Learning<br/><a rel="nofollow noopener" href="https://courses.edx.org/courses/course-v1:Microsoft+DAT256x+1T2018a/course/" target="_blank">https://courses.edx.org/courses/course-v1:Microsoft+DAT256x+1T2018a...</a><br/><a rel="nofollow noopener" href="https://www.khanacademy.org/math/linear-algebra" target="_blank">https://www.khanacademy.org/math/linear-algebra</a><br/><a rel="nofollow noopener" href="https://www.khanacademy.org/math/multivariable-calculus" target="_blank">https://www.khanacademy.org/math/multivariable-calculus</a><br/><br/>You can spend years teaching yourself math, statistics, ML and DL. Pick a simple project and learn these concepts as a means, your learning will go far.<br/><br/>Good Luck!</p>
</div>
</div>
</blockquote> Exactly. I was also stuck. It…tag:www.datasciencecentral.com,2018-05-17:6448529:Comment:7224592018-05-17T21:46:58.821ZRavi Krishnappahttps://www.datasciencecentral.com/profile/RaviKrishnappa
<p>Exactly. I was also stuck. It just shows the poor skills these professors have in writing textbooks. We have to think about newer technologies. One good example is there. <a href="https://www.youtube.com/playlist?list=PLZHQObOWTQDNU6R1_67000Dx_ZCJB-3pi" target="_blank" rel="noopener">3Blue1Brown</a></p>
<p>Exactly. I was also stuck. It just shows the poor skills these professors have in writing textbooks. We have to think about newer technologies. One good example is there. <a href="https://www.youtube.com/playlist?list=PLZHQObOWTQDNU6R1_67000Dx_ZCJB-3pi" target="_blank" rel="noopener">3Blue1Brown</a></p> Here is what I recommend when…tag:www.datasciencecentral.com,2018-05-09:6448529:Comment:7209062018-05-09T16:16:54.085ZAnil Kumarhttps://www.datasciencecentral.com/profile/NeilKumar
<p>Here is what I recommend when I'm mentoring and had good success with it<br></br><br></br>Ian GoodFellow is fairly advanced.<br></br><br></br>I would suggest you start with Grokking Deep Learning by Andrew Trask.<br></br>This book has the lowest barrier to entry possible in author's own words and I can attest he has achieved his mission.<br></br><a href="https://www.manning.com/books/grokking-deep-learning" target="_blank">https://www.manning.com/books/grokking-deep-learning</a><br></br>He breaks down all the concepts…</p>
<p>Here is what I recommend when I'm mentoring and had good success with it<br/><br/>Ian GoodFellow is fairly advanced.<br/><br/>I would suggest you start with Grokking Deep Learning by Andrew Trask.<br/>This book has the lowest barrier to entry possible in author's own words and I can attest he has achieved his mission.<br/><a href="https://www.manning.com/books/grokking-deep-learning" target="_blank">https://www.manning.com/books/grokking-deep-learning</a><br/>He breaks down all the concepts into extremely small and simple Intuition.<br/>you will be doing Gradient Descent & Back Propagation in your sleep :)<br/><br/>After that, I would read <span>Deep Learning with Python by François Chollet, creator of Keras<br/></span><a href="https://www.manning.com/books/deep-learning-with-python" target="_blank">https://www.manning.com/books/deep-learning-with-python</a><br/>Now you will be armed with DNN, ConvNet, RNN, GAN, and ML(Linear Regression, Logistic & Kernel Methods). <br/><br/>For Math, here are some resources I'm listing in relevance to Deep Learning<br/><a href="https://courses.edx.org/courses/course-v1:Microsoft+DAT256x+1T2018a/course/" target="_blank">https://courses.edx.org/courses/course-v1:Microsoft+DAT256x+1T2018a/course/</a><br/><a href="https://www.khanacademy.org/math/linear-algebra" target="_blank">https://www.khanacademy.org/math/linear-algebra</a><br/><a href="https://www.khanacademy.org/math/multivariable-calculus" target="_blank">https://www.khanacademy.org/math/multivariable-calculus</a><br/><br/>You can spend years teaching yourself math, statistics, ML and DL. Pick a simple project and learn these concepts as a means, your learning will go far.<br/><br/>Good Luck!</p> Matrix product is basic to un…tag:www.datasciencecentral.com,2017-10-25:6448529:Comment:6413102017-10-25T13:35:14.864ZAbenezer Girmahttps://www.datasciencecentral.com/profile/AbenezerGirma
<p>Matrix product is basic to understand ML. I think this may help you to visualize what is happening. In addition, follow the suggestion given in above comments.</p>
<p><a href="https://storage.ning.com/topology/rest/1.0/file/get/2773307980?profile=original" target="_self"><img src="https://storage.ning.com/topology/rest/1.0/file/get/2773307980?profile=original" class="align-full" height="126" width="305"/></a></p>
<p></p>
<p>Matrix product is basic to understand ML. I think this may help you to visualize what is happening. In addition, follow the suggestion given in above comments.</p>
<p><a href="https://storage.ning.com/topology/rest/1.0/file/get/2773307980?profile=original" target="_self"><img src="https://storage.ning.com/topology/rest/1.0/file/get/2773307980?profile=original" class="align-full" height="126" width="305"/></a></p>
<p></p> I'd suggest some background i…tag:www.datasciencecentral.com,2017-09-11:6448529:Comment:6193712017-09-11T06:37:37.431ZNaveen Mathew Nathan Shttps://www.datasciencecentral.com/profile/NaveenMathewNathanS
I'd suggest some background in machine learning and neural networks before you start reading the book.<br />
<br />
1) Linear algebra is a must have!<br />
2) Look into the history of neural networks. Start with perceptron and feed forward network with 1 hidden layer before you move onto other architectures - they are fancy, but learning the limitations of perceptron, feed forward networks will truly inspire you to read more.<br />
3) Learn how to interpret weights of neural networks. Hidden layer weights may seem…
I'd suggest some background in machine learning and neural networks before you start reading the book.<br />
<br />
1) Linear algebra is a must have!<br />
2) Look into the history of neural networks. Start with perceptron and feed forward network with 1 hidden layer before you move onto other architectures - they are fancy, but learning the limitations of perceptron, feed forward networks will truly inspire you to read more.<br />
3) Learn how to interpret weights of neural networks. Hidden layer weights may seem insignificant, but they tell you exactly what/how the network learns.<br />
<br />
IMHO these 3 are necessary to understand why other architectures are required and the type of problems that each architecture can solve. I admit that deep learning is a beast, but it can be tamed by using a systematic approach. Ideally, go through a course on deep learning (there are many in YouTube) and use the book as primary reference material. Greetings Frederick,
What th…tag:www.datasciencecentral.com,2017-09-10:6448529:Comment:6192912017-09-10T17:31:06.409ZAvi Silterrahttps://www.datasciencecentral.com/profile/AviSilterra
<p>Greetings Frederick,</p>
<p></p>
<p>What the product operation line is saying (equation 2.5) is the <strong>cell C_{i,j}</strong> is defined as the <strong>dot product of A's row i</strong> and <strong>B's column j</strong>.</p>
<p>A How To video is <a href="https://www.khanacademy.org/math/precalculus/precalc-matrices/multiplying-matrices-by-matrices/v/matrix-multiplication-intro">Intro to matrix multiplication (video) | Khan Academy</a>. This is just the tip of the iceberg for linear…</p>
<p>Greetings Frederick,</p>
<p></p>
<p>What the product operation line is saying (equation 2.5) is the <strong>cell C_{i,j}</strong> is defined as the <strong>dot product of A's row i</strong> and <strong>B's column j</strong>.</p>
<p>A How To video is <a href="https://www.khanacademy.org/math/precalculus/precalc-matrices/multiplying-matrices-by-matrices/v/matrix-multiplication-intro">Intro to matrix multiplication (video) | Khan Academy</a>. This is just the tip of the iceberg for linear algebra. As other commentators noted, a refreshing class on linear algebra might be helpful.</p>
<p>For a mathematical reason why the definition of matrix multiplication is the way it is, see <a href="https://math.stackexchange.com/questions/271927/why-historically-do-we-multiply-matrices-as-we-do" target="_blank">https://math.stackexchange.com/questions/271927/why-historically-do-we-multiply-matrices-as-we-do.</a></p> There are really some great b…tag:www.datasciencecentral.com,2017-09-08:6448529:Comment:6187612017-09-08T17:53:02.864ZAdrian Thompsonhttps://www.datasciencecentral.com/profile/AdrianThompson
<p>There are really some great books I love Mathematics - From the Birth of Numbers which is brilliant in that you get great scope & history. Also I highly value Schaum's outlines series : Linear Algebra (Fully solved problems) walks you thru certain problems and how to solve, then gives you problems to solve, you give it a spin and the answers are provided so you can verify accordingly. I think we also have to give a shout out to the for Dummies series which are are written be great…</p>
<p>There are really some great books I love Mathematics - From the Birth of Numbers which is brilliant in that you get great scope & history. Also I highly value Schaum's outlines series : Linear Algebra (Fully solved problems) walks you thru certain problems and how to solve, then gives you problems to solve, you give it a spin and the answers are provided so you can verify accordingly. I think we also have to give a shout out to the for Dummies series which are are written be great people with great imaginations to convey a thought process in a fun manner.</p>
<p>Get the the library and have some fun ! </p>
<p></p> A good starting point would b…tag:www.datasciencecentral.com,2017-09-06:6448529:Comment:6178322017-09-06T07:50:44.154ZAndreas Wöhrlhttps://www.datasciencecentral.com/profile/AndreasWoehrl
<p>A good starting point would be <a href="https://www.coursera.org/learn/datasciencemathskills" target="_blank">https://www.coursera.org/learn/datasciencemathskills</a> also this one doesn't cover matrices as far as I remember. Andrew NG's Machine Learning course (<a href="https://www.coursera.org/learn/machine-learning/home/week/1" target="_blank">https://www.coursera.org/learn/machine-learning/home/week/1</a>) also has a linear algebra section in week 1 to refresh your knowledge.</p>
<p>A good starting point would be <a href="https://www.coursera.org/learn/datasciencemathskills" target="_blank">https://www.coursera.org/learn/datasciencemathskills</a> also this one doesn't cover matrices as far as I remember. Andrew NG's Machine Learning course (<a href="https://www.coursera.org/learn/machine-learning/home/week/1" target="_blank">https://www.coursera.org/learn/machine-learning/home/week/1</a>) also has a linear algebra section in week 1 to refresh your knowledge.</p> You may want to take a course…tag:www.datasciencecentral.com,2017-09-03:6448529:Comment:6167652017-09-03T11:43:00.169ZNicholas Halehttps://www.datasciencecentral.com/profile/NicholasHale
<p>You may want to take a course in Linear Algebra before getting into ML. It won't make a lot of sense unless you have a basic understanding of matrices. </p>
<p>You may want to take a course in Linear Algebra before getting into ML. It won't make a lot of sense unless you have a basic understanding of matrices. </p>