Summercamp
time
ML
knowledge
Day 0
Day 1
Machine learning without a PhD
- Simple NN (acc. 92%)
- Multilayered NN (acc 95%)
- Optimized MNN (acc 99.5%)
Day 1
time
ML
knowledge
Day 2
Machine learning hands-on
- Weights/Biases
- Activator functions
- Cost functions
- Cross entropy
Day 2
Machine learning hands-on
Day 2
time
Hands-on
ML
knowledge
Day 3
Open AI
- Free data
- Easy reward measurements
- Implement your networks within
the "gym" of openAI - Compete with others and help
ML to progress
Day 3
OpenAI programmer that introduced:
- OpenAI Gym
- OpenAI Universe
- Sources to start learning
(as a software developer)
Day 3
time
Hands-on
ML
knowledge
Day 3/4
Day 4
Linear algebra
- Scalars, Vectors, Matrices, Tensors
- Addition, Multiplication
- Linear dependent/independent
Day 4
Day 4
time
Hands-on
ML
knowledge
Day 4
time
Math
knowledge
Day 5
Machine learning
by Stanford University
Day 5
Machine learning
by Stanford University
- What is ML?
- Supervised
- Unsupervised
- Model representation
- Cost function
- Etc..
Day 5
time
Hands-on
ML
knowledge
Day 6
time
Hands-on
ML
knowledge
+
Conclusion
- ML is hard
- Literature is key
- Experienced colleagues can help a lot
- Discussing with colleagues helps as well
Future?