-
@@ -222,7 +224,7 @@ MathJax.Hub.Config({
@@ -200,7 +201,7 @@ MathJax.Hub.Config({
-Reinforcement learning is a sub-area of machine learning. It studies -how agents take actions based on trial and error, so as to maximize -some notion of cumulative reward in a dynamic system or -environment. Due to its generality, the problem has also been studied -in many other disciplines, such as game theory, control theory, -operations research, information theory, multi-agent systems, swarm -intelligence, statistics, and genetic algorithms. +This is an important topic if we aim at extracting a probability +distribution. This gives us also a confidence interval and error +estimates.
-In March 2016, AlphaGo, a computer program that plays the board game -Go, beat Lee Sedol in a five-game match. This was the first time a -computer Go program had beaten a 9-dan (highest rank) professional -without handicaps. AlphaGo is based on deep convolutional neural -networks and reinforcement learning. AlphaGo’s victory was a major -milestone in artificial intelligence and it has also made -reinforcement learning a hot research area in the field of machine -learning. +Bayesian machine learning allows us to encode our prior beliefs about +what those models should look like, independent of what the data tells +us. This is especially useful when we don’t have a ton of data to +confidently learn our model.
@@ -226,7 +220,7 @@ learning.
-The goal of transfer learning is to transfer the model or knowledge -obtained from a source task to the target task, in order to resolve -the issues of insufficient training data in the target task. The -rationality of doing so lies in that usually the source and target -tasks have inter-correlations, and therefore either the features, -samples, or models in the source task might provide useful information -for us to better solve the target task. Transfer learning is a hot -research topic in recent years, with many problems still waiting to be -solved in this space. +Reinforcement learning is a sub-area of machine learning. It studies +how agents take actions based on trial and error, so as to maximize +some notion of cumulative reward in a dynamic system or +environment. Due to its generality, the problem has also been studied +in many other disciplines, such as game theory, control theory, +operations research, information theory, multi-agent systems, swarm +intelligence, statistics, and genetic algorithms. + +
+In March 2016, AlphaGo, a computer program that plays the board game +Go, beat Lee Sedol in a five-game match. This was the first time a +computer Go program had beaten a 9-dan (highest rank) professional +without handicaps. AlphaGo is based on deep convolutional neural +networks and reinforcement learning. AlphaGo’s victory was a major +milestone in artificial intelligence and it has also made +reinforcement learning a hot research area in the field of machine +learning.
@@ -218,7 +228,7 @@ solved in this space.
-The conventional deep generative model has a potential problem: the -model tends to generate extreme instances to maximize the -probabilistic likelihood, which will hurt its performance. Adversarial -learning utilizes the adversarial behaviors (e.g., generating -adversarial instances or training an adversarial model) to enhance the -robustness of the model and improve the quality of the generated -data. In recent years, one of the most promising unsupervised learning -technologies, generative adversarial networks (GAN), has already been -successfully applied to image, speech, and text. +The goal of transfer learning is to transfer the model or knowledge +obtained from a source task to the target task, in order to resolve +the issues of insufficient training data in the target task. The +rationality of doing so lies in that usually the source and target +tasks have inter-correlations, and therefore either the features, +samples, or models in the source task might provide useful information +for us to better solve the target task. Transfer learning is a hot +research topic in recent years, with many problems still waiting to be +solved in this space.
@@ -218,7 +220,7 @@ successfully applied to image, speech, and text.
-Dual learning is a new learning paradigm, the basic idea of which is -to use the primal-dual structure between machine learning tasks to -obtain effective feedback/regularization, and guide and strengthen the -learning process, thus reducing the requirement of large-scale labeled -data for deep learning. The idea of dual learning has been applied to -many problems in machine learning, including machine translation, -image style conversion, question answering and generation, image -classification and generation, text classification and generation, -image-to-text, and text-to-image. +The conventional deep generative model has a potential problem: the +model tends to generate extreme instances to maximize the +probabilistic likelihood, which will hurt its performance. Adversarial +learning utilizes the adversarial behaviors (e.g., generating +adversarial instances or training an adversarial model) to enhance the +robustness of the model and improve the quality of the generated +data. In recent years, one of the most promising unsupervised learning +technologies, generative adversarial networks (GAN), has already been +successfully applied to image, speech, and text.
@@ -218,7 +220,7 @@ image-to-text, and text-to-image.
-Distributed computation will speed up machine learning algorithms, -significantly improve their efficiency, and thus enlarge their -application. When distributed meets machine learning, more than just -implementing the machine learning algorithms in parallel is required. +Dual learning is a new learning paradigm, the basic idea of which is +to use the primal-dual structure between machine learning tasks to +obtain effective feedback/regularization, and guide and strengthen the +learning process, thus reducing the requirement of large-scale labeled +data for deep learning. The idea of dual learning has been applied to +many problems in machine learning, including machine translation, +image style conversion, question answering and generation, image +classification and generation, text classification and generation, +image-to-text, and text-to-image.
@@ -213,7 +220,7 @@ implementing the machine learning algorithms in parallel is required.
-Meta learning is an emerging research direction in machine -learning. Roughly speaking, meta learning concerns learning how to -learn, and focuses on the understanding and adaptation of the learning -itself, instead of just completing a specific learning task. That is, -a meta learner needs to be able to evaluate its own learning methods -and adjust its own learning methods according to specific learning -tasks. +Distributed computation will speed up machine learning algorithms, +significantly improve their efficiency, and thus enlarge their +application. When distributed meets machine learning, more than just +implementing the machine learning algorithms in parallel is required.
@@ -215,6 +214,8 @@ tasks.
-While there has been much progress in machine learning, there are also challenges. - -
-For example, the mainstream machine learning technologies are -black-box approaches, making us concerned about their potential -risks. To tackle this challenge, we may want to make machine learning -more explainable and controllable. As another example, the -computational complexity of machine learning algorithms is usually -very high and we may want to invent lightweight algorithms or -implementations. Furthermore, in many domains such as physics, -chemistry, biology, and social sciences, people usually seek elegantly -simple equations (e.g., the Schrödinger equation) to uncover the -underlying laws behind various phenomena. In the field of machine -learning, can we reveal simple laws instead of designing more complex -models for data fitting? Although there are many challenges, we are -still very optimistic about the future of machine learning. As we look -forward to the future, here are what we think the research hotspots in -the next ten years will be. +Meta learning is an emerging research direction in machine +learning. Roughly speaking, meta learning concerns learning how to +learn, and focuses on the understanding and adaptation of the learning +itself, instead of just completing a specific learning task. That is, +a meta learner needs to be able to evaluate its own learning methods +and adjust its own learning methods according to specific learning +tasks.
@@ -225,6 +216,7 @@ the next ten years will be.
-Machine learning, especially deep learning, evolves rapidly. The -ability gap between machine and human on many complex cognitive tasks -becomes narrower and narrower. However, we are still in the very early -stage in terms of explaining why those effective models work and how -they work. +While there has been much progress in machine learning, there are also challenges.
-What is missing: the gap between correlation and causation Most -machine learning techniques, especially the statistical ones, depend -highly on data correlation to make predictions and analyses. In -contrast, rational humans tend to reply on clear and trustworthy -causality relations obtained via logical reasoning on real and clear -facts. It is one of the core goals of explainable machine learning to -transition from solving problems by data correlation to solving -problems by logical reasoning. +For example, the mainstream machine learning technologies are +black-box approaches, making us concerned about their potential +risks. To tackle this challenge, we may want to make machine learning +more explainable and controllable. As another example, the +computational complexity of machine learning algorithms is usually +very high and we may want to invent lightweight algorithms or +implementations. Furthermore, in many domains such as physics, +chemistry, biology, and social sciences, people usually seek elegantly +simple equations (e.g., the Schrödinger equation) to uncover the +underlying laws behind various phenomena. In the field of machine +learning, can we reveal simple laws instead of designing more complex +models for data fitting? Although there are many challenges, we are +still very optimistic about the future of machine learning. As we look +forward to the future, here are what we think the research hotspots in +the next ten years will be.
@@ -221,6 +226,7 @@ problems by logical reasoning.
-Quantum machine learning is an emerging interdisciplinary research -area at the intersection of quantum computing and machine learning. +Machine learning, especially deep learning, evolves rapidly. The +ability gap between machine and human on many complex cognitive tasks +becomes narrower and narrower. However, we are still in the very early +stage in terms of explaining why those effective models work and how +they work.
-Quantum computers use effects such as quantum coherence and quantum -entanglement to process information, which is fundamentally different -from classical computers. Quantum algorithms have surpassed the best -classical algorithms in several problems (e.g., searching for an -unsorted database, inverting a sparse matrix), which we call quantum -acceleration. - -
-When quantum computing meets machine learning, it can be a mutually -beneficial and reinforcing process, as it allows us to take advantage -of quantum computing to improve the performance of classical machine -learning algorithms. In addition, we can also use the machine learning -algorithms (on classic computers) to analyze and improve quantum -computing systems. +What is missing: the gap between correlation and causation Most +machine learning techniques, especially the statistical ones, depend +highly on data correlation to make predictions and analyses. In +contrast, rational humans tend to reply on clear and trustworthy +causality relations obtained via logical reasoning on real and clear +facts. It is one of the core goals of explainable machine learning to +transition from solving problems by data correlation to solving +problems by logical reasoning.
@@ -223,6 +222,7 @@ computing systems.
-Many quantum machine learning algorithms are based on variants of -quantum algorithms for solving linear equations, which can efficiently -solve N-variable linear equations with complexity of O(log2 N) under -certain conditions. The quantum matrix inversion algorithm can -accelerate many machine learning methods, such as least square linear -regression, least square version of support vector machine, Gaussian -process, and more. The training of these algorithms can be simplified -to solve linear equations. The key bottleneck of this type of quantum -machine learning algorithms is data input—that is, how to initialize -the quantum system with the entire data set. Although efficient -data-input algorithms exist for certain situations, how to efficiently -input data into a quantum system is as yet unknown for most cases. +Quantum machine learning is an emerging interdisciplinary research +area at the intersection of quantum computing and machine learning. + +
+Quantum computers use effects such as quantum coherence and quantum +entanglement to process information, which is fundamentally different +from classical computers. Quantum algorithms have surpassed the best +classical algorithms in several problems (e.g., searching for an +unsorted database, inverting a sparse matrix), which we call quantum +acceleration. + +
+When quantum computing meets machine learning, it can be a mutually +beneficial and reinforcing process, as it allows us to take advantage +of quantum computing to improve the performance of classical machine +learning algorithms. In addition, we can also use the machine learning +algorithms (on classic computers) to analyze and improve quantum +computing systems.
@@ -216,6 +224,7 @@ input data into a quantum system is as yet unknown for most cases.
-In quantum reinforcement learning, a quantum agent interacts with the -classical environment to obtain rewards from the environment, so as to -adjust and improve its behavioral strategies. In some cases, it -achieves quantum acceleration by the quantum processing capabilities -of the agent or the possibility of exploring the environment through -quantum superposition. Such algorithms have been proposed in -superconducting circuits and systems of trapped ions. +Many quantum machine learning algorithms are based on variants of +quantum algorithms for solving linear equations, which can efficiently +solve N-variable linear equations with complexity of O(log2 N) under +certain conditions. The quantum matrix inversion algorithm can +accelerate many machine learning methods, such as least square linear +regression, least square version of support vector machine, Gaussian +process, and more. The training of these algorithms can be simplified +to solve linear equations. The key bottleneck of this type of quantum +machine learning algorithms is data input—that is, how to initialize +the quantum system with the entire data set. Although efficient +data-input algorithms exist for certain situations, how to efficiently +input data into a quantum system is as yet unknown for most cases.
@@ -210,6 +217,7 @@ superconducting circuits and systems of trapped ions.
-Dedicated quantum information processors, such as quantum annealers -and programmable photonic circuits, are well suited for building deep -quantum networks. The simplest deep quantum network is the Boltzmann -machine. The classical Boltzmann machine consists of bits with tunable -interactions and is trained by adjusting the interaction of these bits -so that the distribution of its expression conforms to the statistics -of the data. To quantize the Boltzmann machine, the neural network can -simply be represented as a set of interacting quantum spins that -correspond to an adjustable Ising model. Then, by initializing the -input neurons in the Boltzmann machine to a fixed state and allowing -the system to heat up, we can read out the output qubits to get the -result. +In quantum reinforcement learning, a quantum agent interacts with the +classical environment to obtain rewards from the environment, so as to +adjust and improve its behavioral strategies. In some cases, it +achieves quantum acceleration by the quantum processing capabilities +of the agent or the possibility of exploring the environment through +quantum superposition. Such algorithms have been proposed in +superconducting circuits and systems of trapped ions.
@@ -214,6 +211,7 @@ result.
-Machine learning aims to imitate how humans -learn. While we have developed successful machine learning algorithms, -until now we have ignored one important fact: humans are social. Each -of us is one part of the total society and it is difficult for us to -live, learn, and improve ourselves, alone and isolated. Therefore, we -should design machines with social properties. Can we let machines -evolve by imitating human society so as to achieve more effective, -intelligent, interpretable “social machine learning”? - -
-And much more. +Dedicated quantum information processors, such as quantum annealers +and programmable photonic circuits, are well suited for building deep +quantum networks. The simplest deep quantum network is the Boltzmann +machine. The classical Boltzmann machine consists of bits with tunable +interactions and is trained by adjusting the interaction of these bits +so that the distribution of its expression conforms to the statistics +of the data. To quantize the Boltzmann machine, the neural network can +simply be represented as a set of interacting quantum spins that +correspond to an adjustable Ising model. Then, by initializing the +input neurons in the Boltzmann machine to a fixed state and allowing +the system to heat up, we can read out the output qubits to get the +result.
@@ -212,6 +215,7 @@ And much more.
-Early computer scientist Alan Kay said, The best way to predict the -future is to create it. Therefore, all machine learning -practitioners, whether scholars or engineers, professors or students, -need to work together to advance these important research -topics. Together, we will not just predict the future, but create it. +Machine learning aims to imitate how humans +learn. While we have developed successful machine learning algorithms, +until now we have ignored one important fact: humans are social. Each +of us is one part of the total society and it is difficult for us to +live, learn, and improve ourselves, alone and isolated. Therefore, we +should design machines with social properties. Can we let machines +evolve by imitating human society so as to achieve more effective, +intelligent, interpretable “social machine learning”? + +
+And much more.
@@ -205,6 +213,7 @@ topics. Together, we will not just predict the future, but create it.
-
diff --git a/doc/pub/summary/html/summary-bs.html b/doc/pub/summary/html/summary-bs.html index 0cfad85cf..b6fe4a094 100644 --- a/doc/pub/summary/html/summary-bs.html +++ b/doc/pub/summary/html/summary-bs.html @@ -73,28 +73,29 @@ Automatically generated HTML file from DocOnce source '___sec14'), ('Additional courses of interest', 2, None, '___sec15'), ("What's the future like?", 2, None, '___sec16'), - ('Reinforcement Learning', 2, None, '___sec17'), - ('Transfer learning', 2, None, '___sec18'), - ('Adversarial learning', 2, None, '___sec19'), - ('Dual learning', 2, None, '___sec20'), - ('Distributed machine learning', 2, None, '___sec21'), - ('Meta learning', 2, None, '___sec22'), - ('The Challenges Facing Machine Learning', 2, None, '___sec23'), - ('Explainable machine learning', 2, None, '___sec24'), - ('Quantum machine learning', 2, None, '___sec25'), + ('Bayesian Machine Learning', 2, None, '___sec17'), + ('Reinforcement Learning', 2, None, '___sec18'), + ('Transfer learning', 2, None, '___sec19'), + ('Adversarial learning', 2, None, '___sec20'), + ('Dual learning', 2, None, '___sec21'), + ('Distributed machine learning', 2, None, '___sec22'), + ('Meta learning', 2, None, '___sec23'), + ('The Challenges Facing Machine Learning', 2, None, '___sec24'), + ('Explainable machine learning', 2, None, '___sec25'), + ('Quantum machine learning', 2, None, '___sec26'), ('Quantum machine learning algorithms based on linear algebra', 2, None, - '___sec26'), - ('Quantum reinforcement learning', 2, None, '___sec27'), - ('Quantum deep learning', 2, None, '___sec28'), - ('Social machine learning', 2, None, '___sec29'), - ('The last words?', 2, None, '___sec30'), + '___sec27'), + ('Quantum reinforcement learning', 2, None, '___sec28'), + ('Quantum deep learning', 2, None, '___sec29'), + ('Social machine learning', 2, None, '___sec30'), + ('The last words?', 2, None, '___sec31'), ('Best wishes to you all and thanks so much for your heroic ' 'efforts this semester', 2, None, - '___sec31')]} + '___sec32')]} end of tocinfo --> @@ -149,21 +150,22 @@ MathJax.Hub.Config({
-
@@ -222,7 +224,7 @@ MathJax.Hub.Config({
-
@@ -161,7 +161,6 @@ MathJax.Hub.Config({
What? Me worry? No final exam in this course!
-
+This is an important topic if we aim at extracting a probability +distribution. This gives us also a confidence interval and error +estimates. + +
+Bayesian machine learning allows us to encode our prior beliefs about +what those models should look like, independent of what the data tells +us. This is especially useful when we don’t have a ton of data to +confidently learn our model. +
Reinforcement learning is a sub-area of machine learning. It studies
@@ -479,7 +493,7 @@ learning.
The goal of transfer learning is to transfer the model or knowledge
@@ -495,7 +509,7 @@ solved in this space.
The conventional deep generative model has a potential problem: the
@@ -511,7 +525,7 @@ successfully applied to image, speech, and text.
Dual learning is a new learning paradigm, the basic idea of which is
@@ -527,7 +541,7 @@ image-to-text, and text-to-image.
Distributed computation will speed up machine learning algorithms,
@@ -538,7 +552,7 @@ implementing the machine learning algorithms in parallel is required.
Meta learning is an emerging research direction in machine
@@ -552,7 +566,7 @@ tasks.
While there has been much progress in machine learning, there are also challenges.
@@ -577,7 +591,7 @@ the next ten years will be.
Machine learning, especially deep learning, evolves rapidly. The
@@ -599,7 +613,7 @@ problems by logical reasoning.
Quantum machine learning is an emerging interdisciplinary research
@@ -624,7 +638,7 @@ computing systems.
Many quantum machine learning algorithms are based on variants of
@@ -643,7 +657,7 @@ input data into a quantum system is as yet unknown for most cases.
In quantum reinforcement learning, a quantum agent interacts with the
@@ -657,7 +671,7 @@ superconducting circuits and systems of trapped ions.
Dedicated quantum information processors, such as quantum annealers
@@ -676,7 +690,7 @@ result.
Machine learning aims to imitate how humans
@@ -694,7 +708,7 @@ And much more.
Early computer scientist Alan Kay said, The best way to predict the
@@ -706,7 +720,7 @@ topics. Together, we will not just predict the future, but create it.
-
+This is an important topic if we aim at extracting a probability
+distribution. This gives us also a confidence interval and error
+estimates.
+
+
+Bayesian machine learning allows us to encode our prior beliefs about
+what those models should look like, independent of what the data tells
+us. This is especially useful when we don’t have a ton of data to
+confidently learn our model.
+
+
+
Reinforcement learning is a sub-area of machine learning. It studies
@@ -437,7 +452,7 @@ learning.
The goal of transfer learning is to transfer the model or knowledge
@@ -453,7 +468,7 @@ solved in this space.
The conventional deep generative model has a potential problem: the
@@ -469,7 +484,7 @@ successfully applied to image, speech, and text.
Dual learning is a new learning paradigm, the basic idea of which is
@@ -485,7 +500,7 @@ image-to-text, and text-to-image.
Distributed computation will speed up machine learning algorithms,
@@ -496,7 +511,7 @@ implementing the machine learning algorithms in parallel is required.
Meta learning is an emerging research direction in machine
@@ -510,7 +525,7 @@ tasks.
While there has been much progress in machine learning, there are also challenges.
@@ -535,7 +550,7 @@ the next ten years will be.
Machine learning, especially deep learning, evolves rapidly. The
@@ -557,7 +572,7 @@ problems by logical reasoning.
Quantum machine learning is an emerging interdisciplinary research
@@ -582,7 +597,7 @@ computing systems.
Many quantum machine learning algorithms are based on variants of
@@ -601,7 +616,7 @@ input data into a quantum system is as yet unknown for most cases.
In quantum reinforcement learning, a quantum agent interacts with the
@@ -615,7 +630,7 @@ superconducting circuits and systems of trapped ions.
Dedicated quantum information processors, such as quantum annealers
@@ -634,7 +649,7 @@ result.
Machine learning aims to imitate how humans
@@ -652,7 +667,7 @@ And much more.
Early computer scientist Alan Kay said, The best way to predict the
@@ -664,7 +679,7 @@ topics. Together, we will not just predict the future, but create it.
-
+This is an important topic if we aim at extracting a probability
+distribution. This gives us also a confidence interval and error
+estimates.
+
+
+Bayesian machine learning allows us to encode our prior beliefs about
+what those models should look like, independent of what the data tells
+us. This is especially useful when we don’t have a ton of data to
+confidently learn our model.
+
+
+
Reinforcement learning is a sub-area of machine learning. It studies
@@ -442,7 +457,7 @@ learning.
The goal of transfer learning is to transfer the model or knowledge
@@ -458,7 +473,7 @@ solved in this space.
The conventional deep generative model has a potential problem: the
@@ -474,7 +489,7 @@ successfully applied to image, speech, and text.
Dual learning is a new learning paradigm, the basic idea of which is
@@ -490,7 +505,7 @@ image-to-text, and text-to-image.
Distributed computation will speed up machine learning algorithms,
@@ -501,7 +516,7 @@ implementing the machine learning algorithms in parallel is required.
Meta learning is an emerging research direction in machine
@@ -515,7 +530,7 @@ tasks.
While there has been much progress in machine learning, there are also challenges.
@@ -540,7 +555,7 @@ the next ten years will be.
Machine learning, especially deep learning, evolves rapidly. The
@@ -562,7 +577,7 @@ problems by logical reasoning.
Quantum machine learning is an emerging interdisciplinary research
@@ -587,7 +602,7 @@ computing systems.
Many quantum machine learning algorithms are based on variants of
@@ -606,7 +621,7 @@ input data into a quantum system is as yet unknown for most cases.
In quantum reinforcement learning, a quantum agent interacts with the
@@ -620,7 +635,7 @@ superconducting circuits and systems of trapped ions.
Dedicated quantum information processors, such as quantum annealers
@@ -639,7 +654,7 @@ result.
Machine learning aims to imitate how humans
@@ -657,7 +672,7 @@ And much more.
Early computer scientist Alan Kay said, The best way to predict the
@@ -669,7 +684,7 @@ topics. Together, we will not just predict the future, but create it.
Transfer learning
+Transfer learning
Adversarial learning
+Adversarial learning
Dual learning
+Dual learning
Distributed machine learning
+Distributed machine learning
Meta learning
+Meta learning
The Challenges Facing Machine Learning
+The Challenges Facing Machine Learning
Explainable machine learning
+Explainable machine learning
Quantum machine learning
+Quantum machine learning
Quantum machine learning algorithms based on linear algebra
+Quantum machine learning algorithms based on linear algebra
Quantum reinforcement learning
+Quantum reinforcement learning
Quantum deep learning
+Quantum deep learning
Social machine learning
+Social machine learning
The last words?
+The last words?
Best wishes to you all and thanks so much for your heroic efforts this semester
+Best wishes to you all and thanks so much for your heroic efforts this semester
diff --git a/doc/pub/summary/html/summary-solarized.html b/doc/pub/summary/html/summary-solarized.html
index 58e3b5571..23faa7654 100644
--- a/doc/pub/summary/html/summary-solarized.html
+++ b/doc/pub/summary/html/summary-solarized.html
@@ -67,28 +67,29 @@ div { text-align: justify; text-justify: inter-word; }
'___sec14'),
('Additional courses of interest', 2, None, '___sec15'),
("What's the future like?", 2, None, '___sec16'),
- ('Reinforcement Learning', 2, None, '___sec17'),
- ('Transfer learning', 2, None, '___sec18'),
- ('Adversarial learning', 2, None, '___sec19'),
- ('Dual learning', 2, None, '___sec20'),
- ('Distributed machine learning', 2, None, '___sec21'),
- ('Meta learning', 2, None, '___sec22'),
- ('The Challenges Facing Machine Learning', 2, None, '___sec23'),
- ('Explainable machine learning', 2, None, '___sec24'),
- ('Quantum machine learning', 2, None, '___sec25'),
+ ('Bayesian Machine Learning', 2, None, '___sec17'),
+ ('Reinforcement Learning', 2, None, '___sec18'),
+ ('Transfer learning', 2, None, '___sec19'),
+ ('Adversarial learning', 2, None, '___sec20'),
+ ('Dual learning', 2, None, '___sec21'),
+ ('Distributed machine learning', 2, None, '___sec22'),
+ ('Meta learning', 2, None, '___sec23'),
+ ('The Challenges Facing Machine Learning', 2, None, '___sec24'),
+ ('Explainable machine learning', 2, None, '___sec25'),
+ ('Quantum machine learning', 2, None, '___sec26'),
('Quantum machine learning algorithms based on linear algebra',
2,
None,
- '___sec26'),
- ('Quantum reinforcement learning', 2, None, '___sec27'),
- ('Quantum deep learning', 2, None, '___sec28'),
- ('Social machine learning', 2, None, '___sec29'),
- ('The last words?', 2, None, '___sec30'),
+ '___sec27'),
+ ('Quantum reinforcement learning', 2, None, '___sec28'),
+ ('Quantum deep learning', 2, None, '___sec29'),
+ ('Social machine learning', 2, None, '___sec30'),
+ ('The last words?', 2, None, '___sec31'),
('Best wishes to you all and thanks so much for your heroic '
'efforts this semester',
2,
None,
- '___sec31')]}
+ '___sec32')]}
end of tocinfo -->
@@ -130,14 +131,13 @@ MathJax.Hub.Config({
Nov 27, 2019
Nov 28, 2019
What? Me worry? No final exam in this course!
-
@@ -174,7 +174,6 @@ The following topics will be covered
-Reinforcement Learning
+Bayesian Machine Learning
+
+
+
+Reinforcement Learning
-Transfer learning
+Transfer learning
-Adversarial learning
+Adversarial learning
-Dual learning
+Dual learning
-Distributed machine learning
+Distributed machine learning
-Meta learning
+Meta learning
-The Challenges Facing Machine Learning
+The Challenges Facing Machine Learning
-Explainable machine learning
+Explainable machine learning
-Quantum machine learning
+Quantum machine learning
-Quantum machine learning algorithms based on linear algebra
+Quantum machine learning algorithms based on linear algebra
-Quantum reinforcement learning
+Quantum reinforcement learning
-Quantum deep learning
+Quantum deep learning
-Social machine learning
+Social machine learning
-The last words?
+The last words?
-Best wishes to you all and thanks so much for your heroic efforts this semester
+Best wishes to you all and thanks so much for your heroic efforts this semester
diff --git a/doc/pub/summary/html/summary.html b/doc/pub/summary/html/summary.html
index 188ab35ba..e5a9adca1 100644
--- a/doc/pub/summary/html/summary.html
+++ b/doc/pub/summary/html/summary.html
@@ -72,28 +72,29 @@ div { text-align: justify; text-justify: inter-word; }
'___sec14'),
('Additional courses of interest', 2, None, '___sec15'),
("What's the future like?", 2, None, '___sec16'),
- ('Reinforcement Learning', 2, None, '___sec17'),
- ('Transfer learning', 2, None, '___sec18'),
- ('Adversarial learning', 2, None, '___sec19'),
- ('Dual learning', 2, None, '___sec20'),
- ('Distributed machine learning', 2, None, '___sec21'),
- ('Meta learning', 2, None, '___sec22'),
- ('The Challenges Facing Machine Learning', 2, None, '___sec23'),
- ('Explainable machine learning', 2, None, '___sec24'),
- ('Quantum machine learning', 2, None, '___sec25'),
+ ('Bayesian Machine Learning', 2, None, '___sec17'),
+ ('Reinforcement Learning', 2, None, '___sec18'),
+ ('Transfer learning', 2, None, '___sec19'),
+ ('Adversarial learning', 2, None, '___sec20'),
+ ('Dual learning', 2, None, '___sec21'),
+ ('Distributed machine learning', 2, None, '___sec22'),
+ ('Meta learning', 2, None, '___sec23'),
+ ('The Challenges Facing Machine Learning', 2, None, '___sec24'),
+ ('Explainable machine learning', 2, None, '___sec25'),
+ ('Quantum machine learning', 2, None, '___sec26'),
('Quantum machine learning algorithms based on linear algebra',
2,
None,
- '___sec26'),
- ('Quantum reinforcement learning', 2, None, '___sec27'),
- ('Quantum deep learning', 2, None, '___sec28'),
- ('Social machine learning', 2, None, '___sec29'),
- ('The last words?', 2, None, '___sec30'),
+ '___sec27'),
+ ('Quantum reinforcement learning', 2, None, '___sec28'),
+ ('Quantum deep learning', 2, None, '___sec29'),
+ ('Social machine learning', 2, None, '___sec30'),
+ ('The last words?', 2, None, '___sec31'),
('Best wishes to you all and thanks so much for your heroic '
'efforts this semester',
2,
None,
- '___sec31')]}
+ '___sec32')]}
end of tocinfo -->
@@ -135,14 +136,13 @@ MathJax.Hub.Config({
Nov 27, 2019
Nov 28, 2019
What? Me worry? No final exam in this course!
-
@@ -179,7 +179,6 @@ The following topics will be covered
-Reinforcement Learning
+Bayesian Machine Learning
+
+
+
+Reinforcement Learning
-Transfer learning
+Transfer learning
-Adversarial learning
+Adversarial learning
-Dual learning
+Dual learning
-Distributed machine learning
+Distributed machine learning
-Meta learning
+Meta learning
-The Challenges Facing Machine Learning
+The Challenges Facing Machine Learning
-Explainable machine learning
+Explainable machine learning
-Quantum machine learning
+Quantum machine learning
-Quantum machine learning algorithms based on linear algebra
+Quantum machine learning algorithms based on linear algebra
-Quantum reinforcement learning
+Quantum reinforcement learning
-Quantum deep learning
+Quantum deep learning
-Social machine learning
+Social machine learning
-The last words?
+The last words?
-Best wishes to you all and thanks so much for your heroic efforts this semester
+Best wishes to you all and thanks so much for your heroic efforts this semester
diff --git a/doc/pub/summary/ipynb/ipynb-summary-src.tar.gz b/doc/pub/summary/ipynb/ipynb-summary-src.tar.gz
index c190be9d4..fe5e116d2 100644
Binary files a/doc/pub/summary/ipynb/ipynb-summary-src.tar.gz and b/doc/pub/summary/ipynb/ipynb-summary-src.tar.gz differ
diff --git a/doc/pub/summary/ipynb/summary.ipynb b/doc/pub/summary/ipynb/summary.ipynb
index a5654f7cf..6eec7c647 100644
--- a/doc/pub/summary/ipynb/summary.ipynb
+++ b/doc/pub/summary/ipynb/summary.ipynb
@@ -10,7 +10,7 @@
" \n",
"**Morten Hjorth-Jensen Email morten.hjorth-jensen@fys.uio.no**, Department of Physics and Center of Mathematics for Applications, University of Oslo and National Superconducting Cyclotron Laboratory, Michigan State University\n",
"\n",
- "Date: **Nov 27, 2019**\n",
+ "Date: **Nov 28, 2019**\n",
"\n",
"Copyright 1999-2019, Morten Hjorth-Jensen Email morten.hjorth-jensen@fys.uio.no. Released under CC Attribution-NonCommercial 4.0 license\n",
"\n",
@@ -26,14 +26,6 @@
"\n",
"\n",
"\n",
- "\n",
- "\n",
- "\n",
- "\n",
- "![](\"figures/whatmeworry.jpeg\")
\n",
- "\n",
- "\n",
- "\n",
"\n",
"\n",
"## What did I learn in school this year?\n",
@@ -66,17 +58,15 @@
"\n",
"2. Simpler models, binomial distribution, the Poisson distribution, simple and multivariate normal distributions;\n",
"\n",
- "3. Central elements of Bayesian statistics and modeling;\n",
+ "3. Central elements from linear algebra\n",
"\n",
- "4. Central elements from linear algebra\n",
+ "4. Gradient methods for data optimization\n",
"\n",
- "5. Gradient methods for data optimization\n",
+ "5. Estimation of errors using cross-validation, blocking, bootstrapping and jackknife methods;\n",
"\n",
- "6. Estimation of errors using cross-validation, blocking, bootstrapping and jackknife methods;\n",
+ "6. Practical optimization using Singular-value decomposition and least squares for parameterizing data.\n",
"\n",
- "7. Practical optimization using Singular-value decomposition and least squares for parameterizing data.\n",
- "\n",
- "8. Principal Component Analysis.\n",
+ "7. Principal Component Analysis.\n",
"\n",
"## Machine learning\n",
"\n",
@@ -275,6 +265,20 @@
"\n",
"3. Encoder-decoder framework, which is mostly used for image or sequence generation, such as machine translation, text summarization, and image captioning.\n",
"\n",
+ "## Bayesian Machine Learning\n",
+ "\n",
+ "This is an important topic if we aim at extracting a probability\n",
+ "distribution. This gives us also a confidence interval and error\n",
+ "estimates.\n",
+ "\n",
+ "Bayesian machine learning allows us to encode our prior beliefs about\n",
+ "what those models should look like, independent of what the data tells\n",
+ "us. This is especially useful when we don’t have a ton of data to\n",
+ "confidently learn our model.\n",
+ "\n",
+ "\n",
+ "\n",
+ "\n",
"## Reinforcement Learning\n",
"\n",
"Reinforcement learning is a sub-area of machine learning. It studies\n",
diff --git a/doc/pub/summary/pdf/summary-minted.pdf b/doc/pub/summary/pdf/summary-minted.pdf
index fc6a35364..0e04406cf 100644
Binary files a/doc/pub/summary/pdf/summary-minted.pdf and b/doc/pub/summary/pdf/summary-minted.pdf differ
diff --git a/doc/src/Summary/summary.do.txt b/doc/src/Summary/summary.do.txt
index 96d40f960..a7e6a78fb 100644
--- a/doc/src/Summary/summary.do.txt
+++ b/doc/src/Summary/summary.do.txt
@@ -6,7 +6,6 @@ DATE: today
!split
===== What? Me worry? No final exam in this course! =====
FIGURE: [figures/exam1.jpeg, width=500 frac=0.6]
-FIGURE: [figures/whatmeworry.jpeg, width=500 frac=0.6]
!split
@@ -32,7 +31,6 @@ o Machine learning
The following topics will be covered
o Basic concepts, expectation values, variance, covariance, correlation functions and errors;
o Simpler models, binomial distribution, the Poisson distribution, simple and multivariate normal distributions;
-o Central elements of Bayesian statistics and modeling;
o Central elements from linear algebra
o Gradient methods for data optimization
o Estimation of errors using cross-validation, blocking, bootstrapping and jackknife methods;
@@ -201,6 +199,22 @@ o Convolutional neural networks, which are mostly used in image and video data p
o Recurrent neural networks, which can process sequential data of variable length and have been widely used in natural language understanding and speech processing;
o Encoder-decoder framework, which is mostly used for image or sequence generation, such as machine translation, text summarization, and image captioning.
+
+!split
+===== Bayesian Machine Learning =====
+
+This is an important topic if we aim at extracting a probability
+distribution. This gives us also a confidence interval and error
+estimates.
+
+Bayesian machine learning allows us to encode our prior beliefs about
+what those models should look like, independent of what the data tells
+us. This is especially useful when we don’t have a ton of data to
+confidently learn our model.
+
+
+
+
!split
===== Reinforcement Learning =====