<html><head></head><body style="overflow-wrap: break-word; -webkit-nbsp-mode:
space; line-break: after-white-space;"><div>## (Q1)
Definitions</div><div><br></div><div>**i) Intelligence:**
</div><div>Intelligence is the capacity for abstraction, logic,
understanding, self-awareness, learning, reasoning, planning, creativity, critical
thinking, and problem-solving. It enables an entity to perceive or infer
information and use it to adaptively respond to its
environment[2].</div><div><br></div><div>**ii) Artificial Intelligence:**
</div><div>Artificial intelligence (AI) is the intelligence demonstrated by
machines or computers that enables them to imitate or mimic human capabilities such
as perception, reasoning, learning, and
problem-solving[3].</div><div><br></div><div>**iii) Agent:** </div><div>An
agent in AI is a software program or entity that interacts with its environment,
collects data, and uses this data to autonomously perform tasks to achieve
predetermined goals[4].</div><div><br></div><div>---</div><div><br></div><div>##
(Q2) Uninformed Search Algorithms</div><div><br></div><div>- **Breadth-First Search
(BFS):** </div><div> BFS explores all nodes at the present depth before
moving to the next level. It uses a queue and guarantees finding the shortest path
in an unweighted graph but can consume a lot of
memory[5].</div><div><br></div><div>- **Uniform-Cost Search (UCS):**
</div><div> UCS expands the node with the lowest path cost. It uses a
priority queue and is optimal for finding the least-cost path in a weighted graph,
but can be slower than BFS for large graphs[6].</div><div><br></div><div>- **Depth-
First Search (DFS):** </div><div> DFS explores as far as possible along
each branch before backtracking. It uses a stack and is memory efficient but may
get stuck in infinite branches if the graph is deep or
infinite[7].</div><div><br></div><div>- **Iterative Deepening Search:**
</div><div> This combines the space efficiency of DFS with the
optimality of BFS by performing DFS to increasing depth limits until the goal is
found. It is useful when the search space is large and the depth of the solution is
unknown.</div><div><br></div><div>- **Bidirectional Search:**
</div><div> Bidirectional search runs two simultaneous searches-one
forward from the start and one backward from the goal-until they meet. It can be
much faster than unidirectional search if both directions can be efficiently
explored[8].</div><div><br></div><div>---</div><div><br></div><div>## (Q3) Task
Environment Specification</div><div><br></div><div>When designing an agent, the
first step is to specify the task environment as completely as possible. This
includes defining the environment’s properties (observable, deterministic,
episodic, static, discrete), the agent’s possible actions, percepts, goals, and
performance measures. For example, in a chess game, the environment is fully
observable, deterministic, episodic, static, and discrete. In contrast, a self-
driving car operates in a partially observable, stochastic, sequential, dynamic,
and continuous environment. Clear specification ensures the agent is designed
appropriately for its intended
context.</div><div><br></div><div>---</div><div><br></div><div>## (Q4) Speech
Processing & Human vs. Computer</div><div><br></div><div>**a. Goal of Speech
Processing Research in AI:** </div><div>The goal is to enable computers to
understand, interpret, and generate human language, facilitating natural and
effective communication between humans and machines.</div><div><br></div><div>**b.
Comparison: Computer vs. Human (Tabular Form)**</div><div><br></div><div>| Area
| Computer Strengths
| Human
Strengths
|</div><div>|---------------------|-------------------------------------------|----
-------------------------------------|</div><div>| Speed
| Processes data rapidly
| Slower processing
|</div><div>| Accuracy
| High precision in calculations
| Prone to errors
|</div><div>| Creativity
| Limited, follows programmed logic
| Highly creative and innovative |
</div><div>| Learning | Learns from large
datasets (machine learning) | Learns from experience and context
|</div><div>| Adaptability | Needs reprogramming
for new tasks | Adapts quickly to new situations
|</div><div><br></div><div>**c. PEAS for Taxi
Driving**</div><div><br></div><div>- **Performance Measure:** Safe driving, timely
arrival, passenger satisfaction, fuel efficiency</div><div>- **Environment:**
Roads, traffic signals, pedestrians, weather conditions</div><div>- **Actuators:**
Steering, accelerator, brake, indicator, horn</div><div>- **Sensors:** Cameras,
GPS, speedometer, microphone,
lidar/radar</div><div><br></div><div>---</div><div><br></div><div>## (Q5) BFS vs.
DFS & Uniform Cost Search Application</div><div><br></div><div>**a. Differences
Between BFS and DFS (Tabular Form)**</div><div><br></div><div>| Feature
| Breadth-First Search (BFS)
| Depth-First Search (DFS)
|</div><div>|-------------------|------------------------------------|-------
--------------------------|</div><div>| Memory Usage | High
(stores all nodes at a level) | Low (stores only current path) |</div><div>|
Path Found | Shortest path in unweighted graphs | May
not find shortest path |</div><div><br></div><div>**b. Uniform
Cost Search on Figure 1**</div><div><br></div><div>Applying UCS to the tree in
Figure 1, the goal nodes are reached in order of increasing path cost. The order
and cost for each goal node are:</div><div><br></div><div>| Goal Node | Cost
|</div><div>|-----------|------|</div><div>| T | 1
|</div><div>| E | 2
|</div><div>| A | 3 |</div><div>| R
| 3 |</div><div>| S
| 3 |</div><div>| F | 4
|</div><div>| I | 4 |</div><div>| N
| 5 |</div><div>| D
| 6 |</div><div>| L | 8
|</div><div><br></div><div>This order shows how UCS always expands the
lowest-cost node available, ensuring optimality for path
cost[6].</div><div></div></body></html>
