Ai PPT New

Reinforcement
Learning
Presented by:
-Baljit kaur
-Loveleen kaur
-Kulwinder singh
-Lovepreet kaur
One subfield of machine learning is reinforcement
learning. It involves acting appropriately to
maximize reward in a certain circumstance. It is
utilized by a variety of software and devices to
determine the optimal course of action or behavior
for a given circumstance. Reinforcement learning
and supervised learning are different in that in
supervised learning, the model is trained with the
correct answer already present in the training data,
while in reinforcement learning, the model is
trained without an answer and is guided by the
reinforcement agent's decision on how to complete
the task at hand.
A self-governing, self-teaching system,
reinforcement learning basically learns by making
mistakes. It learns by doing in order to attain the
best results, or, to put it another way, it takes
actions with the intention of maximizing rewards.
For instance: The issue is this: There are numerous
obstacles between our agent and the reward. The
agent's job is to determine the most efficient route
to the reward. For easier understanding, consider
the problem that follows..
In reinforcement learning, creators come up with
a way to penalize bad conduct and reward good
behaviour. With this approach, the agent is
encouraged to use the desired activities by
assigning them positive values, while the
undesirable behaviours are discouraged by
assigning them negative values. In order to find
the best answer, this programs the agent to look
for long-term and maximum total rewards . By
setting long-term objectives, the agent can avoid
becoming bogged down in smaller, less crucial
ones. The agent eventually learns to look for the
positive and steer clear of the negative. Artificial
intelligence (AI) has embraced this learning
technique to guide unsupervised machine learning
using incentives, or positive reinforcement, and
punishments, or negative reinforcement.
Two categories of reinforcement exist:
Positive: Favorable Reinforcement is the process through
which an event that results from a specific behavior
intensifies and repeats the behavior. Put otherwise, it
produces a favorable impact on behavior. Among
reinforcement learning's benefits are: Optimizes
Performance Long-term Change Sustain ChangeAn excess of
reinforcement may result in an overabundance of states,
which could reduce the effectiveness.
Negative: Adverse The definition of reinforcement is when a

behavior gets stronger as a result of a bad circumstance
being avoided or halted. Reinforcement learning benefits
include: Boosts ConductShow disdain for a minimal level of
performanceIt only offers enough to satisfy the bare
minimum of conduct..
Policy: The policy defines how an agent behaves at any given moment. Formally, it’s a mapping from
environmental states to specific actions that the agent should take when in those states. Policies can be simple
functions, lookup tables, or involve more complex computations. Importantly, the policy alone determines the
agent’s behavior, and it often involves stochastic choices.
Reward Signal: After each action taken by the agent, the environment provides a single numerical reward. This
reward depends on the agent’s action and the current state. The agent’s ultimate goal is to maximize the total
reward received over the long term. The reward signal guides the agent by indicating what decisions are
beneficial (high reward) or detrimental (low reward).
Value Function: While the reward signal informs the agent about immediate goodness or badness, the value
function looks ahead to the long run. It specifies what is advantageous over extended periods. Essentially, it
helps the agent evaluate the desirability of different states or state-action pairs. The value function
complements the reward signal by considering the cumulative impact of actions over time1.
Environment Model (Optional): Although not always present, an environment model can be useful. It represents
the agent’s understanding of how the environment behaves. This model allows the agent to simulate potential
Playing games: DeepMind's AlphaGo is a well-known example. It has
vanquished world champions in the game of Go by using reinforcement learning.
Reinforcement learning has also been applied to the mastery of other games,
such as Atari games and chess . Robotics: Reinforcement learning is widely
applied in robotics to accomplish tasks like object handling, navigation, and arm
control. In order to interact with their environment and complete tasks quickly,
robots learn by making mistakes . Healthcare: Personalized treatment
recommendations, hospital resource allocation, and clinical trial optimization
are just a few of the activities in which reinforcement learning is used
Reinforcement learning can be applied to tasks like text summarization,

machine translation, and conversation synthesis in natural language processing
(NLP). From user interactions or environmental feedback, the agent learns to
produce coherent and contextually relevant replies . These illustrations show
how adaptable reinforcement learning is in a variety of contexts, where agents
are trained to make decisions by interacting with their surroundings in order to
accomplish predetermined goals.
Reinforcement learning can be used to solve very complex
problems that cannot be solved by conventional techniques.
The model can correct the errors that occurred during the training
process.
In RL, training data is obtained via the direct interaction of the

agent with the environment
Reinforcement learning can handle environments that are non-

deterministic, meaning that the outcomes of actions are not always
predictable. This is useful in real-world applications where the
environment may change over time or is uncertain.
Reinforcement learning can be used to solve a wide range of

problems, including those that involve decision making, control, and
optimization.
Reinforcement learning is a flexible approach that can be combined

with other machine learning techniques, such as deep learning
The difficulties of using reinforcement learningAlthough
it has a lot of potential, reinforcement learning has significant
drawbacks. Its application is still restricted and its
deployment can be challenging. This kind of machine
learning's reliance on environment exploration is one of its
deployment's obstacles.A robot that relies on reinforcement
learning to negotiate a complicated physical world, for
instance, would seek out new states and change its course as
it advances.This method can be computationally costly and
have limited effectiveness due to the amount of time needed
to achieve proper learning. The demands on time and
computational resources increase in tandem with the
complexity of the training environment.
Typical algorithms for reinforcement learningReinforcement learning is not limited to a
single algorithm; rather, it encompasses a variety of methods with somewhat diverse
methodologies. The primary cause of the variations is the various methods they employ
to investigate their surroundings:Reward-state-action-state-action. The first step in
this reinforcement learning process is to provide the agent with a policy. The agent is
not given any policies; instead, it discovers the worth of an activity by investigating its
surroundings. This strategy is more self-directed than model-based. Python
programming is frequently used to write Q-learning implementations in the real
world.Q-networks at depth. These algorithms employ reinforcement learning methods
along with neural networks and deep Q-learning. They employ the self-directed
reinforcement learning environment and go by the name "deep reinforcement learning"
as well.
There are numerous benefits to using reinforcement learning while tackling challenging jobs and
challenges.
Flexibility: In dynamic and unpredictable contexts, reinforcement learning enables agents to adjust and
gain knowledge from past experiences. Because of its adaptability, reinforcement learning (RL) can be
used in a variety of contexts where the environment may change or be unknown.
Autonomy:Once trained, reinforcement learning agents possess the ability to function independently,
including decision-making and action-taking without human guidance. This autonomy is very helpful for
applications like gaming, robotics, and driverless cars.
Adaptability: Real-time agents have the ability to modify their tactics and actions in response to
environmental input. They are able to improve their performance repeatedly by learning from both
positive reinforcement and negative consequences.
1: Reinforcement learning is not preferable to use for solving simple problems.

2: Reinforcement learning needs a lot of data and a lot of computation
3: Reinforcement learning is highly dependent on the quality of the reward function. If the
reward function is poorly designed, the agent may not learn the desired behavior.
4: Reinforcement learning can be difficult to debug and interpret. It is not always clear why
the agent is behaving in a certain way, which can make it difficult to diagnose and fix
problems.
supervised learning, algorithms In unsupervised learning, Reinforcement learning. This
train on a body of labeled data. developers turn algorithms loose takes a different approach. It
Supervised learning algorithms on fully unlabeled data. The situates an agent in an
can only learn attributes that algorithms learn by cataloging their environment with clear
are specified in the data set. A own observations about data parameters defining beneficial
common application of features without being told what to activity and nonbeneficial activity
supervised learning is image look for. and an overarching endgame to
recognition models. These reach.
models receive a set of labeled
images and learn to distinguish
common attributes of
predefined forms.
Reinforcement
use one or morelearning is projectedincluding
training approaches, to play reinforcement
a bigger rolelearning.
in the future of AI. The other
approaches to training machine learning algorithms require large amounts of
preexisting training data. Reinforcement learning agents, on the other hand, require
the time to gradually learn how to operate via interactions with their environments.
Despite the challenges, various industries are expected to continue exploring
reinforcement learning's potential.Reinforcement learning has already demonstrated
promise in various areas. For example, marketing and advertising firms are using
algorithms trained this way for recommendation engines. Manufacturers are using
reinforcement learning to train their next-generation robotic systems.Scientists at
Alphabet's AI subsidiary, Google DeepMind, have proposed that reinforcement
learning could bring the current state of AI -- often called narrow AI -- to its theoretical
final form of artificial general intelligence. They believe machines that learn through
reinforcement learning will eventually become sentient and operate independently of
human supervision.Machine learning algorithms
Thanks!
Do you have any
questions?

Ai PPT New

Uploaded by

Copyright:

Available Formats

Ai PPT New

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ai PPT New

Uploaded by

Copyright:

Available Formats

Reinforcement

Negative: Adverse The definition of reinforcement is when a

Reinforcement learning can be applied to tasks like text summarization,

In RL, training data is obtained via the direct interaction of the

Reinforcement learning can handle environments that are non-

Reinforcement learning can be used to solve a wide range of

Reinforcement learning is a flexible approach that can be combined

1: Reinforcement learning is not preferable to use for solving simple problems.

You might also like