PENGANTAR MACHINE LEARNING (Pra Kuliah Umum) Betha Nurina Sari,M.Kom
Machine Learning • Machine Learning is the study of computer algorithms that improve automatically through experience. Applications range from data mining programs that discover general rules in large data sets, to information filtering systems that automatically learn users' interests (Tom Mitchell, 1997).
Machine Learning • Machine Learning adalah salah satu disiplin ilmu dari Computer Science yang mempelajari bagaimana membuat komputer/mesin itu mempunyai suatu kecerdasan. Agar mempunyai suatu kecerdasan, komputer/mesin harus dapat belajar.
Machine Learning • Machine Learning adalah suatu bidang keilmuan yang berisi tentang pembelajaran komputer/mesin untuk menjadi cerdas
Traditional Programming VS Machine Learning Traditional Programming
Data Program
Computer
Output
Computer
Program
Machine Learning
Data Output
Magic? No, more like gardening • • • •
Seeds = Algorithms Nutrients = Data Gardener = You Plants = Programs
Sekilas tentang Machine Learning
A Few Quotes • “A breakthrough in machine learning would be worth ten Microsofts” (Bill Gates, Chairman, Microsoft) • “Machine learning is the next Internet” (Tony Tether, Director, DARPA)
• Machine learning is the hot new thing” (John Hennessy, President, Stanford)
• “Web rankings today are mostly a matter of machine learning” (Prabhakar Raghavan, Dir. Research, Yahoo) • “Machine learning is going to result in a real revolution” (Greg Papadopoulos, CTO, Sun)
• “Machine learning is today’s discontinuity” (Jerry Yang, CEO, Yahoo)
Resources: Datasets • UCI Repository: http://www.ics.uci.edu/~mlearn/MLRepository.html • UCI KDD Archive: http://kdd.ics.uci.edu/summary.data.application.html
• Statlib: http://lib.stat.cmu.edu/ • Delve: http://www.cs.utoronto.ca/~delve/
9
Resources: Journals • Journal of Machine Learning Research www.jmlr.org • Machine Learning • IEEE Transactions on Neural Networks • IEEE Transactions on Pattern Analysis and Machine Intelligence • Annals of Statistics • Journal of the American Statistical Association • ... 10
The steps needed for any model-based machine learning application:
Sumber : http://www.mbmlbook.com/LifeCycle.html
Contoh Manfaat Machine Learning
….dst
Topik pembahasan dalam Machine Learning (1) • Probability and Estimation – – – –
Bayes rule Maximum Likelihood Estimation Max a posteriori Naive bayes
• Gaussian Naive Bayes and Logistic Regression – – – –
Gaussian Bayes Classifier Document Classification Form of Decision Surfaces Linear Regression
Topik pembahasan dalam Machine Learning (2) • Graphical Models – Bayes Nets – Conditional Independence – D-Separation and Conditional Indepence – Inference – Learning from fully observed data – Learning from partially observed data – EM algorithm – Mixture of Gaussian Clustering
Topik pembahasan dalam Machine Learning (3) • Hidden Markov Model – Markov models – HMM and Bayes Net – Other probabilistic time series
• Artificial Neural Networks – Non-Linear Regression – Backpropagation and Gradient Descent – Learning Hidden Layer Representation
Topik pembahasan dalam Machine Learning (4) • Learning Representation – PCA
- ICA
- Fisher Linear Discrimination
• Kernel Methods and SVM – Kernels – Maximizing Margin
- SVMs -Noise and Soft Margin
Topik pembahasan dalam Machine Learning (5) • Deep Learning – Early Work – Stacked Auto Decoders
- Why Deep Learning - Deep Belief Networks
Referensi • Slide CSE 446 Machine Learning oleh Pedro Domingos • Slide Introduction to Machine Learning oleh Entin Martiana (2013) • BRP Kuliah Pembelajaran Mesin oleh Ito Wasito (2015)
TUGAS KULIAH UMUM : RESUME • TUGAS RESUME untuk semua mahasiswa yang mengambil mata kuliah sistem pakar, baik yang hadir atau tidak hadir kuliah umum. • Tugas bersifat individu, kalau ditemukan unsur plagiat (copy paste) maka nilai dibagi dengan hasil tugas yang sama. • Resume berupa ringkasan atas materi yang disampaikan tentang Machine Learning
TUGAS KULIAH UMUM : RESUME • Resume maksimal 1 halaman A4, format softfile (.pdf) dikirim paling lambat Jumat, 7 April 2017 23.59 WIB dengan subyek RESUME_NAMA melalui email ke
[email protected] • Komponen yang harus ada : – Identitas mahasiswa (NPM,Nama,Kelas Asal) (5 poin) – Judul/topik resume (5 poin) – Hasil Resume (90 poin) • Ringkasan materi padat, jelas, ringkas • Inspirasi/motivasi yang didapatkan
TUGAS KULIAH UMUM : RESUME • Keterlambatan pengumpulan tugas per-hari, penalti dikurangi 10 poin • Kesalahan penulisan dalam tugas penalti dikurangi 1 poin/kesalahan • Kesalahan format file maka tidak dikoreksi
• Pertemuan 10 KULIAH UMUM MACHINE LEARNING Selasa,4 April 2017 di Aula UNSIKA
