DADS401 ADVANCED MACHINE LEARNING

Description

SESSION	FEBRUARY – MARCH 2024
PROGRAM	MASTER OF BUSINESS ADMINISTRATION (MBA)
SEMESTER	IV
COURSE CODE & NAME	DADS401-ADVANCED MACHINE LEARNING

 

 

Assignment Set – 1

 

1. (a) Discuss the objective of Time Series Analysis.

(b) Explain the merits and demerits of using Smoothing. 

Ans 1.

(a) Objective of Time Series Analysis

Time Series Analysis (TSA) is a statistical method that analyzes time series data, which are data points indexed in time order. The primary objective of TSA is to develop models that can predict future values based on previously observed values. This analysis is crucial in various fields such as economics, business, finance, science, and engineering where it is essential to forecast future trends from past data.

The goals of TSA can be broadly categorized as follows:

• Understanding and

 

Its Half solved only

Buy Complete from our online store

 

https://smuassignment.in/online-store/

 

MUJ Fully solved assignment available for session FEB/March 2024.

 

Lowest price guarantee with quality.

Charges INR 198 only per assignment. For more information you can get via mail or Whats app also

Mail id is aapkieducation@gmail.com

 

Our website www.smuassignment.in

After mail, we will reply you instant or maximum

1 hour.

Otherwise you can also contact on our

whatsapp no 8791490301.

 

2. (a) Define AR (0), AR (1) and AR (2).

(b) Define ARCH Model. Discuss its usage.

Ans 2.

1. AR Models:

Autoregressive (AR) models are a class of linear models where future values of a variable are assumed to be a linear function of past observations. They are commonly used in time series forecasting to describe time-dependent structures in the data.

AR(0) Model: This is the simplest form of the autoregressive model where the current value of the series is equal to a constant term plus a

 

3. (a) Explain some challenges or limitations we face with Deep Learning.

(b) Explain any three applications of AI in Medical sciences.      

Ans 3.

1. Challenges and Limitations of Deep Learning

Deep learning has transformed many technological fields, delivering impressive performance across a range of complex tasks. However, it faces several challenges and limitations that can impact its effectiveness and applicability:

Data Dependency: Deep learning models require large amounts of data to train effectively. This dependency on big data can be a significant hurdle in fields where data is scarce, sensitive, or expensive to acquire.

Computational Cost:

 

 

Assignment Set – 2

 

 

4. (a) Define Back Propagation.

(b) Describe some applications of ANN.    

Ans 4.

1. Definition and Mechanism of Back Propagation

Back propagation is a fundamental algorithm used for training artificial neural networks (ANN). Essential for supervised learning, it facilitates the minimization of error by adjusting the weights of the network. This method involves two main phases: forward pass and backward pass.

In the forward pass, input data is passed through the network, layer by layer, until the output layer produces its predictions.

 

 

5. (a) Differentiate CNN vs RNN.

(b)Write a short note on max pooling and average pooling?

Ans 5.

1. Differentiating CNN and RNN

Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) are two types of neural network architectures designed for different tasks and data types, reflecting their distinct structural characteristics.

CNNs are primarily used for spatial data processing, such as images and videos. A CNN’s architecture is well-suited to capture spatial hierarchies in data by applying convolutional filters that process data in small receptive fields. These networks typically consist of multiple layers of convolutions that apply filters to the input data to extract features, followed by pooling layers that reduce

 

6. (a)What is Auto-Encoder? Explain its classification.

(b)Describe the types of RL algorithm(s).

Ans 6.

1. Auto-Encoders and Classification

An Auto-Encoder is a type of artificial neural network used primarily for unsupervised learning of efficient codings. The goal of an auto-encoder is to learn a representation (encoding) for a set of data, typically for the purpose of dimensionality reduction. Notably, an auto-encoder is trained to ignore noise and recreate the input at its output.

Structure and Mechanism: An auto-encoder typically consists of two main parts: the encoder and the decoder. The encoder