Quantum Neural Network Based Machine Translator for Hindi to English

Table 2

Comparison of performance measurement of MT system with QNN and traditional neural network.

S. No	Quantum interval (θ)	Epoch (iteration)	Training performance (MSE)	Validation performance (MSE)	Test performance (MSE)
1	θ = 0 is equivalent to ANN	20.04391	0.002027	0.002066	0.002075
2	0.25	15.36727	0.000185	0.000205	0.000203
3	0.5	15.47305	0.000248	0.000273	0.000273
4	0.75	15.28343	0.000231	0.000248	0.000252
5	1 *	15.48303	0.00025	0.00027	0.000272
6	1.25	15.67066	0.000284	0.0003	0.0003
7	1.5	15.64271	0.000296	0.000318	0.000316
8	1.75	15.78643	0.000349	0.000374	0.000375
9	2	15.68663	0.000184	0.000204	0.000209
10	2.25	15.96607	0.000249	0.000273	0.000273
11	2.5	16.27345	0.000256	0.000281	0.000276
12	2.75	16.54092	0.00022	0.000242	0.000242
13	3	16.49301	0.000348	0.000361	0.00037
14	3.25	16.93214	0.000192	0.000214	0.000216
15	3.5	17.76248	0.000294	0.000315	0.000323
16	3.75	17.85429	0.000185	0.000202	0.000207
17	4	18.72056	0.000217	0.000237	0.00024

∗For the value of θ = 1, the optimum trade-off of iteration and the minimum Error (MSE) achieved.