Learning Domain-Independent Deep Representations by Mutual Information Minimization

<div>Tradeoff parameter sensitivity study of <span class="nowrap"><svg height="11.8174pt" id="M126" style="vertical-align:-3.1815pt" version="1.1" viewbox="-0.0498162 -8.6359 13.6974 11.8174" width="13.6974pt" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink"><g transform="matrix(.013,0,0,-0.013,0,0)"><path d="M645 631C614 643 545 666 457 666C215 666 23 519 23 283C23 90 158 -16 337 -16C412 -16 489 2 522 10C543 39 590 127 606 167L580 181C519 89 459 18 348 18C201 18 122 136 122 287C122 464 244 632 435 632C544 632 602 595 608 472L639 475C643 526 645 581 645 631Z"></path></g><g transform="matrix(.0091,0,0,-0.0091,8.619,3.132)"><path d="M414 144C384 79 371 75 317 75H135L276 221C367 316 408 376 408 465C408 570 327 635 237 635C179 635 131 609 100 575L42 494L67 471C94 510 138 565 205 565C277 565 321 517 321 435C321 348 258 270 195 195C146 137 88 81 33 26V0H411C423 44 433 88 446 135L414 144Z"></path></g></svg>.</span> (a) Office-31. (b) ImageCLEF-DA. (c) Email spam. (d) CK+. (e) Amazon.</div>

Computational Intelligence and Neuroscience

fig3

Figure 3

Figure 3: Learning Domain-Independent Deep Representations by Mutual Information Minimization 