Ensemble Models in Machine Learning:
Nischal HP (~nischal) |
Ensemble models help us exploit the power of computing. Ensemble methods aren't new. They form the basis for some extremely powerful machine learning algorithms like random forests and gradient boosting machines. The key point about ensemble is that consensus from diverse models are more reliable than a single source. This poster will showcase how we can combine model outputs from various base models(logistic regression, support vector machines, decision trees, neural networks, etc) to create a stronger/better model output.
The primary goal of the poster is to answer the following questions: 1) Why ensembles produce better output?
2) How ensembles produce better output?
3) When data scales, what's the impact? What are the trade-offs to consider?
4) Can ensemble models eliminate expert domain knowledge?
5) What are the various strategies to create ensemble models?
Basic knowledge on machine learning and usage of python libraries like sklearn and pandas.
Power of Ensembles Poster Deck - https://speakerdeck.com/nischalhp/power-of-ensembles
Bargava Subramanian is a Senior Data Scientist at Cisco Systems, India. He has a Masters in Statistics from University of Maryland, College Park, USA. He is a data geek and an ardent NBA follower. On twitter, he can be reached @bargava
Nischal HP is a Data Engineer at Redmart, India. He has a Masters in Computer Science from BITS, Pilani, India. He lives on music, ardent traveler and a mad soccer fan. On twitter, he can be reached @nischalhp
Bargava - https://speakerdeck.com/bargava
Nischal - https://speakerdeck.com/nischalhp