Scribd
Upload
54 views4 pages

What Are The Advantages and Disadvantages of Deep Learning

Uploaded by

vijaykoppadi09
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
54 views4 pages

What Are The Advantages and Disadvantages of Deep Learning

Uploaded by

vijaykoppadi09
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

What are the advantages and disadvantages of Deep Learning?

Advantages and disadvantages of Deep Learning

S.No Advantages Disadvantages

1 Highly Effective in Complex Tasks involving May require large amounts of data for
large datasets and complex patterns, such as training, which can be both costly and time-
image and speech recognition, natural language consuming to obtain
processing, and game playing

2 Deep Learning models can process both Computationally intensive and often
structured and unstructured data effectively requires specialized hardware, like GPUs or
structured data - tabular data TPUs
unstructured data - images, text, audio

3 Automatically learn and extract relevant Prone to overfitting, where they perform
features from raw data, reducing the need for well on training data but generalize poorly
manual feature engineering or domain expertise on new, unseen data

4 State-of-the-Art Performance in several Hyper-parameter tuning process is


domains surpassing human-level performance, tedious: Finding the right hyperparameters
and setting new benchmarks for deep networks can be challenging and
may require extensive experimentation

5 Wide range of applications across industries , Black box in nature as it is almost


including marketing, finance, customer service, impossible to understand what is happening
virtual assistants, chatbots etc. in hidden layers within deep networks

How does Deep Learning methods compare with traditional Machine


Learning methods?
Within the field of ‘Artificial Intelligence,’ there exists ‘Machine Learning,’ where we use
computer algorithms to make predictions about previously unseen data using patterns and
knowledge learned from existing data. Deep Learning is a subset of Machine Learning, and the
technique that separates it from traditional Machine Learning is the focus on Neural Networks,
systems inspired by the structure and function of the human brain. The term ‘deep’ alludes to
the presence of multiple layers within these Neural Networks. Deep Learning is considered a
more modern technology than many others within Machine Learning.

Artificial Intelligence, Machine Learning and Deep Learning


Part of the reason why Deep Learning is a newcomer is due to its reliance on high-end
computer hardware, which has only become more feasible within the last twenty years. Unlike
traditional Machine Learning, which often runs seamlessly on standard business computers,
Deep Learning algorithms demand large amounts of data and substantially longer training
times.
Deep Learning Algorithms
• Feedforward Neural Networks (FNN)
• Transformer Networks
• Recurrent Neural Networks(RNN)
• Long Short-Term Memory (LSTM)
• Convolutional Neural Networks (CNN)
• Generative Adversarial Networks (GANs)
• Autoencoders
Traditional Machine Learning Algorithms
• Linear Regression
• Logistic Regression
• Decision Trees
• Random Forest
• Gradient Boosting
• Support Vector Machines
• Naive Bayes Classifier
• Principal Component Analysis
• Gaussian Mixture Model
• Discriminant Analysis (LDA/QDA)
• K Nearest Neighbor (KNN)

10-point difference between Deep Learning and Traditional Machine


Learning
SNo. Attribute Deep Learning Traditional Machine Learning Winner

1 Feature Low High DL


Engineering
Automatically extracts Requires manual feature
features engineering

2 Domain Expertise Low High DL

Network automatically Domain knowledge required to


learns features design relevant features

3 Model High Low Depends


Complexity
Employs deep neural Uses simpler models like
networks with many decision trees, linear regression,
layers, allowing it to or support vector machines
capture complex patterns (SVMs) to learn patterns in data
and relationships in data

4 Data requirements High Low Traditional


ML
Requires large amount of Can work well with smaller
data to learn effectively datasets

5 Training time High Low Traditional


ML
Training deep neural Tends to have faster training
networks can be time- times
consuming, especially for
complex architectures

6 Hardware Hi-tech Standard Traditional


Requirements ML
SNo. Attribute Deep Learning Traditional Machine Learning Winner

Often requires specialized Can be implemented on standard


hardware like GPUs or CPUs
TPUs to accelerate training

7 Interpretability of Low Varies Traditional


model prediction ML
More like a black-box Very high interpretability for
model making it models like logistic regression,
challenging to interpret decision trees, lower for models
such as GBM

8 Performance on High High Both


simpler tasks with
low data Process large amounts of Well-suited for a wide range of
requirements data and learn from it, tasks. Can be trained on smaller
delivering excellent results datasets and still achieve good
(eg: fraud performance
detection, default
prediction,
clustering)

9 Performance on High Low DL


complex tasks
Excels in learning complex Model limitations to learn highly
(eg: NLP, Image patterns and achieving complex tasks
recognition, state-of-the-art results
Speech
recognition)

10 Training process More tedious Less tedious Traditional


ML
Due to high data Traditional machine learning
requirements and longer involves smaller datasets and
training times, the training shorter training times, allowing
process and for more flexibility in
hyperparameter tuning in experimenting with a wider
Deep Learning is more range of hyperparameters to
challenging enhance model performance.

Deep Learning vs Traditional Machine Learning


The choice between Deep Learning and Traditional Machine Learning depends on
factors such as the nature of the problem, available data, computational resources, and the level
of interpretability required. Both approaches have their strengths and weaknesses, making them
suitable for different types of tasks. Deep learning is more suitable for complex problems with
large amounts of data, while traditional machine learning is more suitable for simpler problems
with limited data.

You might also like