TWGAN: Twin Discriminator Generative Adversarial Networks

Zhaoyu Zhang,Mengyan Li,Haonian Xie,Jun Yu,Tongliang Liu,Chang Wen Chen

DOI: https://doi.org/10.1109/tmm.2021.3057989

IF: 7.3

2022-01-01

IEEE Transactions on Multimedia

Abstract:Generative Adversarial Networks (GAN) has become more and more popular these years. However, it is difficult to train and suffers from the training instability problem. To tackle this difficulty, this paper proposes a novel approach. Our idea is intuitive but proven to be very useful. In essence, it combines saturating loss and non-saturating loss into the loss function. Thus it will exploit the complementary statistical properties from two kinds of loss functions to effectively improve the training stability. We term our method twin discriminator Generative Adversarial Networks (TWGAN), which, unlike GAN, has a generator and a twin discriminator. The twin discriminator consists of two discriminators with identical architecture and both of them aim to distinguish whether the samples are from real data or fake data. We develop theoretical analysis to show that, given the optimal discriminators, optimizing the generator of TWGAN reduces to minimizing the Kullback-Leibler (KL) divergence between the distribution of generated data (<span class="mjpage"><svg xmlns:xlink="http://www.w3.org/1999/xlink" width="2.514ex" height="2.843ex" style="vertical-align: -1.005ex;" viewBox="0 -791.3 1082.3 1223.9" role="img" focusable="false" xmlns="http://www.w3.org/2000/svg"><g stroke="currentColor" fill="currentColor" stroke-width="0" transform="matrix(1 0 0 -1 0 0)"> <use xlink:href="#MJMATHI-50" x="0" y="0"></use> <use transform="scale(0.707)" xlink:href="#MJMATHI-67" x="908" y="-213"></use></g></svg></span>) and the distribution of real data (<span class="mjpage"><svg xmlns:xlink="http://www.w3.org/1999/xlink" width="5.884ex" height="2.509ex" style="vertical-align: -0.671ex;" viewBox="0 -791.3 2533.2 1080.4" role="img" focusable="false" xmlns="http://www.w3.org/2000/svg"><g stroke="currentColor" fill="currentColor" stroke-width="0" transform="matrix(1 0 0 -1 0 0)"> <use xlink:href="#MJMATHI-50" x="0" y="0"></use> <use transform="scale(0.707)" xlink:href="#MJMATHI-64" x="908" y="-213"></use> <use xlink:href="#MJMATHI-61" x="1112" y="0"></use> <use xlink:href="#MJMATHI-74" x="1642" y="0"></use> <use xlink:href="#MJMATHI-61" x="2003" y="0"></use></g></svg></span>), hence effectively addressing the training instability problem. Extensive experiments on MNIST, Fashion MNIST, CIFAR-10/100 and STL-10 datasets demonstrate that the competitive performance of our TWGAN in generating good quality and diverse samples over baselines. The obtained highest inception score (IS) and lowest Fr <span class="mjpage"><svg xmlns:xlink="http://www.w3.org/1999/xlink" width="6.573ex" height="2.009ex" style="vertical-align: -0.338ex;" viewBox="0 -719.6 2830 865.1" role="img" focusable="false" xmlns="http://www.w3.org/2000/svg"><g stroke="currentColor" fill="currentColor" stroke-width="0" transform="matrix(1 0 0 -1 0 0)"> <use xlink:href="#MJMATHI-61" x="0" y="0"></use> <use xlink:href="#MJMATHI-63" x="529" y="0"></use> <use xlink:href="#MJMATHI-75" x="963" y="0"></use> <use xlink:href="#MJMATHI-74" x="1535" y="0"></use> <use xlink:href="#MJMATHI-65" x="1897" y="0"></use> <use xlink:href="#MJMATHI-65" x="2363" y="0"></use></g></svg></span> chet Inception Distance (FID), compared with other state-of-the-art GANs, show the superiority of our TWGAN.<svg xmlns="http://www.w3.org/2000/svg" style="display: none;"><defs id="MathJax_SVG_glyphs"><path stroke-width="1" id="MJMATHI-50" d="M287 628Q287 635 230 637Q206 637 199 638T192 648Q192 649 194 659Q200 679 203 681T397 683Q587 682 600 680Q664 669 707 631T751 530Q751 453 685 389Q616 321 507 303Q500 302 402 301H307L277 182Q247 66 247 59Q247 55 248 54T255 50T272 48T305 46H336Q342 37 342 35Q342 19 335 5Q330 0 319 0Q316 0 282 1T182 2Q120 2 87 2T51 1Q33 1 33 11Q33 13 36 25Q40 41 44 43T67 46Q94 46 127 49Q141 52 146 61Q149 65 218 339T287 628ZM645 554Q645 567 643 575T634 597T609 619T560 635Q553 636 480 637Q463 637 445 637T416 636T404 636Q391 635 386 627Q384 621 367 550T332 412T314 344Q314 342 395 342H407H430Q542 342 590 392Q617 419 631 471T645 554Z"></path><path stroke-width="1" id="MJMATHI-67" d="M311 43Q296 30 267 15T206 0Q143 0 105 45T66 160Q66 265 143 353T314 442Q361 442 401 394L404 398Q406 401 409 404T418 412T431 419T447 422Q461 422 470 413T480 394Q480 379 423 152T363 -80Q345 -134 286 -169T151 -205Q10 -205 10 -137Q10 -111 28 -91T74 -71Q89 -71 102 -80T116 -111Q116 -121 114 -130T107 -144T99 -154T92 -162L90 -164H91Q101 -167 151 -167Q189 -167 211 -155Q234 -144 254 -122T282 -75Q288 -56 298 -13Q311 35 311 43ZM384 328L380 339Q377 350 375 354T369 368T359 382T346 393T328 402T306 405Q262 405 221 352Q191 313 171 233T151 117Q151 38 213 38Q269 38 323 108L331 118L384 328Z"></path><path stroke-width="1" id="MJMATHI-64" d="M366 683Q367 683 438 688T511 694Q523 694 523 686Q523 679 450 384T375 83T374 68Q374 26 402 26Q411 27 422 35Q443 55 463 131Q469 151 473 152Q475 153 483 153H487H491Q506 153 506 145Q506 140 503 129Q490 79 473 48T445 8T417 -8Q409 -10 393 -10Q359 -10 336 5T306 36L300 51Q299 52 296 50Q294 48 292 46Q233 -10 172 -10Q117 -10 75 30T33 157Q33 205 53 255T101 341Q148 398 195 420T280 442Q336 442 364 400Q369 394 369 396Q370 400 396 505T424 616Q424 629 417 632T378 637H357Q351 643 351 645T353 664Q358 683 366 683ZM352 326Q329 405 277 405Q242 405 210 374T160 293Q131 214 119 129Q119 126 119 118T118 106Q118 61 136 44T179 26Q233 26 290 98L298 109L352 326Z"></path><path stroke-width="1" id="MJMATHI-61" d="M33 157Q33 258 109 349T280 441Q331 441 370 392Q386 422 416 422Q429 422 439 414T449 394Q449 381 412 234T374 68Q374 43 381 35T402 26Q411 27 422 35Q443 55 463 131Q469 151 473 152Q475 153 483 153H487Q506 153 506 144Q506 138 501 117T481 63T449 13Q436 0 417 -8Q409 -10 393 -10Q359 -10 336 5T306 36L300 51Q299 52 296 50Q294 48 292 46Q233 -10 172 -10Q117 -10 75 30T33 157ZM351 328Q351 334 346 350T323 385T277 405Q242 405 210 374T160 293Q131 214 119 129Q119 126 119 118T118 106Q118 61 136 44T179 26Q217 26 254 59T298 110Q300 114 325 217T351 328Z"></path><path stroke-width="1" id="MJMATHI-74" d="M26 385Q19 392 19 395Q19 399 22 411T27 425Q29 430 36 430T87 431H140L159 511Q162 522 166 540T173 566T179 586T187 603T197 615T211 624T229 626Q247 625 254 615T261 596Q261 589 252 549T232 470L222 433Q222 431 272 431H323Q330 424 330 420Q330 398 317 385H210L174 240Q135 80 135 68Q135 26 162 26Q197 26 230 60T283 144Q285 150 288 151T303 153H307Q322 153 322 145Q322 142 319 133Q314 117 301 95T267 48T216 6T155 -11Q125 -11 98 4T59 56Q57 64 57 83V101L92 241Q127 382 128 383Q128 385 77 385H26Z"></path><path stroke-width="1" id="MJMATHI-63" d="M34 159Q34 268 120 355T306 442Q362 442 394 418T427 355Q427 326 408 306T360 285Q341 285 330 295T319 325T330 359T352 380T366 386H367Q367 388 361 392T340 400T306 404Q276 404 249 390Q228 381 206 359Q162 315 142 235T121 119Q121 73 147 50Q169 26 205 26H209Q321 26 394 111Q403 121 406 121Q410 121 419 112T429 98T420 83T391 55T346 25T282 0T202 -11Q127 -11 81 37T34 159Z"></path><path stroke-width="1" id="MJMATHI-75" d="M21 287Q21 295 30 318T55 370T99 420T158 442Q204 442 227 417T250 358Q250 340 216 246T182 105Q182 62 196 45T238 27T291 44T328 78L339 95Q341 99 377 247Q407 367 413 387T427 416Q444 431 463 431Q480 431 488 421T496 402L420 84Q419 79 419 68Q419 43 426 35T447 26Q469 29 482 57T512 145Q514 153 532 153Q551 153 551 144Q550 139 549 130T540 98T523 55T498 17T462 -8Q454 -10 438 -10Q372 -10 347 46Q345 45 336 36T318 21T296 6T267 -6T233 -11Q189 -11 155 7Q103 38 103 113Q103 170 138 262T173 379Q173 380 173 381Q173 390 173 393T169 400T158 404H154Q131 404 112 385T82 344T65 302T57 280Q55 278 41 278H27Q21 284 21 287Z"></path><path stroke-width="1" id="MJMATHI-65" d="M39 168Q39 225 58 272T107 350T174 402T244 433T307 442H310Q355 442 388 420T421 355Q421 265 310 237Q261 224 176 223Q139 223 138 221Q138 219 132 186T125 128Q125 81 146 54T209 26T302 45T394 111Q403 121 406 121Q410 121 419 112T429 98T420 82T390 55T344 24T281 -1T205 -11Q126 -11 83 42T39 168ZM373 353Q367 405 305 405Q272 405 244 391T199 357T170 316T154 280T149 261Q149 260 169 260Q282 260 327 284T373 353Z"></path></defs></svg>