what are some macine learning algorithms that you should always have a strong understanding of and

this airtcle is wrote by  Sean Owen,Director ,Data Science @Cloudera

I think one needs to have a competent knowledge of 1-2 techniques in:

• Regression
• Classification
• Clustering
• Collaborative filtering
• (Bonus) Inference via graphical models

Certainly, it's valuable and important to understand simple  Linear regression .

Gradient descent  is important because it underpins common classifier techniques like  Logistic regression . Also: the  Support vector machine . 
 
I also strongly enc ourage people to have a working knowl edge of  Random forest  classification / regression. It's inherently an ensemble method, effective, and has different properties from the above.

K-means++  clustering is a must.

For  collaborative fi ltering, neigh borhood metho ds are  simple enough that almost don't deserve me n tion. I  would try t o understand  latent fa cto r models based on low-ran k matrix  facto rizati on like the  Singular value decomposition   or simple alternating least squares ( http://yifanhu.net/PUB/cf.pdf )

Bonus: MCMC methods ( Markov chain Monte Carlo ) for graphical models.