1. Question 1 — Applications of Deep Learning (Select all that apply)

• ❌ Speech enactment

• ✅ Color restoration in grayscale images

• ❌ Automatic coding

• ✅ Automatic handwriting generation

Explanation:
Deep learning is widely used for image-to-image tasks and generative tasks like handwriting.

---

2. Question 2 — Components of a Neural Network (Select all that apply)

• ✅ Input layer

• ✅ Hidden layer

• ❌ Sparse layer

• ✅ Output layer

• ❌ Intermediate layer

Explanation:
Standard network components: Input → Hidden → Output.

---

3. Question 3 — Lip-syncing for dubbed videos

• ❌ Color restoration

• ✅ Speech enactment

• ❌ Automatic sound generation

• ❌ Text-to-image

Explanation:
Speech enactment aligns mouth movements with speech — ideal for lip-sync automation.

---

4. Question 4 — Weighted Sum Calculation

Given:
x1=0.3, x2=0.7
w1=0.2, w2=0.4
b=0.1

Compute:
z = x1w1 + x2w2 + b
= (0.3×0.2) + (0.7×0.4) + 0.1
= 0.06 + 0.28 + 0.1
= 0.44

• ✅ 0.440

• ❌ 0.340

• ❌ 0.520

• ❌ 0.380

---

5. Question 5 — Sigmoid Activation

z = –0.4
σ(z) = 1 / (1 + e^0.4) ≈ 1 / (1 + 1.4918) ≈ 0.401

• ❌ 0.350

• ✅ 0.401

• ❌ 0.376

• ❌ 0.450

---

6. Question 6 — Where are signals processed?

• ❌ Synapse

• ❌ Axon

• ❌ Dendrite

• ✅ Soma

Explanation:
Soma = processing unit of the biological neuron.

---

7. Question 7 — Correct Biological Neural Flow

• ✅ Dendrites → Soma → Axon → Synapse

• ❌ Other options

Explanation:
This is the natural order of information flow.

---

8. Question 8 — Primary Purpose of Deep Learning

• ❌ Rule-based systems

• ❌ Only image recognition

• ❌ Replace programming

• ✅ Learn complex patterns from large data automatically

---

9. Question 9 — Correct Layer Connectivity

• ❌ Bidirectional all layers

• ❌ Hidden layers independent

• ❌ Random connectivity

• ✅ Information flows forward: Input → Hidden → Output

---

10. Question 10 — Role of Weights and Biases

• ❌ Control learning rate

• ✅ Weights = connection strength; Biases = shift activation for flexibility

• ❌ Only used for error correction

• ❌ Store final outputs

---

🧾 Summary Table

---