Learning Scenarios( Supervised learning, Unsupervised learning, Semi-

Supervised learning, Transductive inference, On-line learning,

Reinforcement learning, Active learning), Generalization
Supervised Learning, Unsupervised Learning, Reinforcement learning )

Learning is a search through the space of possible hypotheses for one

that will perform well, even on new examples beyond the training set. To
measure the accuracy of a hypothesis we give it a test set of examples
that are distinct from the training set.

By Prof. Ankur Choudhary

Learning Scenarios
Learning Problems
Statistical Inference
1. Supervised Learning 7. Inductive Learning
2. Unsupervised Learning 8. Deductive Inference
3. Reinforcement Learning 9. Transductive Learning
Learning Techniques
Hybrid Learning
10. Multi-Task Learning
Problems
11. Active Learning
4. Semi-Supervised Learning 12. Online Learning
5. Self-Supervised Learning 13. Transfer Learning
6. Multi-Instance Learning 14. Ensemble Learning

By Prof. Ankur Choudhary

 1. Supervised Learning
 “Applications in which the training data comprises examples of the input vectors
along with their corresponding target vectors are known as supervised learning
problems.”
 Supervised Learning is the process of making an algorithm to learn to map
an input to a particular output. This is achieved using the labelled datasets
that you have collected. If the mapping is correct, the algorithm has
successfully learned. Else, you make the necessary changes to the algorithm
so that it can learn correctly. Supervised Learning algorithms can help make
predictions for new unseen data that we obtain later in the future.

By Prof. Ankur Choudhary

 Types of Supervised Learning
 Supervised Learning has been broadly classified into 2 types.
 Regression
 Classification
 Regression is the kind of Supervised Learning that learns from the
Labelled Datasets and is then able to predict a continuous-
valued output for the new data given to the algorithm. It is used
whenever the output required is a number such as money or height
etc.
 Classification, on the other hand, is the kind of learning where the
algorithm needs to map the new data that is obtained to any one of
the 2 classes that we have in our dataset. The classes need to be
mapped to either 1 or 0 which in real-life translated to „Yes‟ or „No‟,
„Rains‟ or „Does Not Rain‟ and so forth. The output will be either
one of the classes and not a number as it was in Regression.

By Prof. Ankur Choudhary

By Prof. Ankur Choudhary
 2. Unsupervised Learning
 In unsupervised learning, there is no instructor or teacher, and the
algorithm must learn to make sense of the data without this guide.
 Compared to supervised learning, unsupervised learning operates
upon only the input data without outputs or target variables. As
such, unsupervised learning does not have a teacher correcting the
model, as in the case of supervised learning.
 There are many types of unsupervised learning, although there are
two main problems that are often encountered by a practitioner:
they are clustering that involves finding groups in the data and
density estimation that involves summarizing the distribution of
data.
 Clustering: Unsupervised learning problem that involves finding
groups in data.
 Density Estimation: Unsupervised learning problem that involves
summarizing the distribution of data.

By Prof. Ankur Choudhary

The most common unsupervised learning task is clustering: detecting potentially useful
clusters of input examples. For example, a taxi agent might gradually develop a concept of
“good traffic days” and “bad traffic days” without ever being given labeled examples of
each by a teacher.

By Prof. Ankur Choudhary

3. Reinforcement Learning
 Reinforcement learning is learning what to do — how
to map situations to actions—so as to maximize a
numerical reward signal. The learner is not told which
actions to take, but instead must discover which
actions yield the most reward by trying them.
 Some machine learning algorithms do not just experience
a fixed dataset. For example, reinforcement learning
algorithms interact with an environment, so there is a
feedback loop between the learning system and its
experiences.

By Prof. Ankur Choudhary

By Prof. Ankur Choudhary
4. Semi-Supervised Learning
 Semi-supervised learning is supervised learning where the training data
contains very few labeled examples and a large number of unlabeled
examples.
 The goal of a semi-supervised learning model is to make effective use of
all of the available data, not just the labelled data like in supervised
learning.
 Making effective use of unlabelled data may require the use of or
inspiration from unsupervised methods such as clustering and density
estimation. Once groups or patterns are discovered, supervised methods
or ideas from supervised learning may be used to label the unlabeled
examples or apply labels to unlabeled representations later used for
prediction.
 “Semisupervised” learning attempts to improve the accuracy of supervised
learning by exploiting information in unlabeled data. This sounds like magic, but
it can work!
 For example, classifying photographs requires a dataset of photographs
that have already been labeled by human operators.

By Prof. Ankur Choudhary

By Prof. Ankur Choudhary
 5. Self-Supervised Learning
 The self-supervised learning framework requires only unlabeled data in order to
formulate a pretext learning task such as predicting context or image rotation,
for which a target objective can be computed without supervision.
 A common example of self-supervised learning is computer vision
where a corpus of unlabeled images is available and can be used to
train a supervised model, such as making images grayscale and
having a model predict a color representation (colorization) or
removing blocks of the image and have a model predict the missing
parts (inpainting).
 Another example of self-supervised learning is generative
adversarial networks, or GANs. These are generative models that
are most commonly used for creating synthetic photographs using
only a collection of unlabeled examples from the target domain.

By Prof. Ankur Choudhary

By Prof. Ankur Choudhary
 6. Multi-Instance Learning
 Multi-instance learning is a supervised learning problem
where individual examples are unlabeled; instead, bags or
groups of samples are labeled.
 In multi-instance learning, an entire collection of examples is
labeled as containing or not containing an example of a
class, but the individual members of the collection are not
labeled.
 Modeling involves using knowledge that one or some of
the instances in a bag are associated with a target label,
and to predict the label for new bags in the future given
their composition of multiple unlabeled examples.

By Prof. Ankur Choudhary

By Prof. Ankur Choudhary
7. Inductive Learning
 Inductive learning involves using evidence to determine
the outcome.
 Inductive reasoning refers to using specific cases to
determine general outcomes, e.g. specific to general.
 However, one disadvantage of the inductive approach is
the risk that the learner will formulate a rule
incorrectly. For this reason, it is important to check that
the learner has inferred the correct rule. Also, if a rule
is complex it may be better to use the deductive
approach and give the rule first, or give some guidance.

By Prof. Ankur Choudhary

By Prof. Ankur Choudhary
 8. Deductive Learning
 Deduction is the reverse of induction. If induction is going
from the specific to the general, deduction is going from
the general to the specific.
 Deduction is a top-down type of reasoning that seeks for
all premises to be met before determining the conclusion,
whereas induction is a bottom-up type of reasoning that
uses available data as evidence for an outcome.

By Prof. Ankur Choudhary

By Prof. Ankur Choudhary
 9.Transductive Learning
 Transduction is reasoning from observed, specific
(training) cases to specific (test) cases. In
contrast, induction is reasoning from observed training
cases to general rules, which are then applied to the test
cases.


By Prof. Ankur Choudhary

10. Multi-Task Learning

 Multi-task learning is a type of supervised learning that involves fitting a

model on one dataset that addresses multiple related problems.
 It involves devising a model that can be trained on multiple related
tasks in such a way that the performance of the model is improved by
training across the tasks as compared to being trained on any single
task.
 Multi-task learning is a way to improve generalization by pooling
the examples (which can be seen as soft constraints imposed on
the parameters) arising out of several tasks.
 For example, it is common for a multi-task learning problem to involve
the same input patterns that may be used for multiple different outputs
or supervised learning problems. In this setup, each output may be
predicted by a different part of the model, allowing the core of the
model to generalize across each task for the same inputs.

By Prof. Ankur Choudhary

 11. Active Learning
 Active learning is a type of supervised learning and seeks to
achieve the same or better performance of so-called “passive”
supervised learning, although by being more efficient about
what data is collected or used by the model.
 The key idea behind active learning is that a machine learning algorithm can achieve
greater accuracy with fewer training labels if it is allowed to choose the data from
which it learns. An active learner may pose queries, usually in the form of unlabeled
data instances to be labeled by an oracle (e.g., a human annotator).

 Active learning is a useful approach when there is not much

data available and new data is expensive to collect or label.

 The active learning process allows the sampling of the domain

to be directed in a way that minimizes the number of samples
and maximizes the effectiveness of the model.

By Prof. Ankur Choudhary

12. Online Learning
 Online learning involves using the data available and
updating the model directly before a prediction is required
or after the last observation was made.
 Online learning is appropriate for those problems where
observations are provided over time and where the
probability distribution of observations is expected to also
change over time. Therefore, the model is expected to
change just as frequently in order to capture and harness
those changes.
 Traditionally machine learning is performed offline, which means we have a
batch of data, and we optimize an equation […] However, if we have
streaming data, we need to perform online learning, so we can update our
estimates as each new data point arrives rather than waiting until “the end”
(which may never occur).
 One example of online learning is so-called stochastic or
online gradient descent used to fit an artificial neural
network.

By Prof. Ankur Choudhary

13. Transfer Learning
 Transfer learning is a type of learning where a model is first
trained on one task, then some or all of the model is used as
the starting point for a related task.

 In transfer learning, the learner must perform two or

more different tasks, but we assume that many of the
factors that explain the variations in P1 are relevant to
the variations that need to be captured for learning P2.
 An example is image classification, where a predictive model,
such as an artificial neural network, can be trained on a large
corpus of general images, and the weights of the model can
be used as a starting point when training on a smaller more
specific dataset, such as dogs and cats. The features already
learned by the model on the broader task, such as extracting
lines and patterns, will be helpful on the new related task.

By Prof. Ankur Choudhary

14. Ensemble Learning
 Ensemble learning is an approach where two or
more modes are fit on the same data and the
predictions from each model are combined.
 The field of ensemble learning provides many ways of combining the ensemble members’
predictions, including uniform weighting and weights chosen on a validation set.

 Bagging, boosting, and stacking have been

developed over the last couple of decades, and
their performance is often astonishingly good.
Machine learning researchers have struggled to
understand why.

By Prof. Ankur Choudhary