Abstract
Web image datasets curated online inherently contain ambiguous in-distribution instances and out-of-distribution instances, which we collectively call
non-conforming
(NC) instances. In many recent approaches for mitigating the negative effects of NC instances, the core implicit assumption is that the NC instances can be found via entropy maximization. For “entropy” to be well-defined, we are interpreting the output prediction vector of an instance as the parameter vector of a multinomial random variable, with respect to some trained model with a softmax output layer. Hence, entropy maximization is based on the idealized assumption that NC instances have predictions that are “almost” uniformly distributed. However, in real-world web image datasets, there are numerous NC instances whose predictions are far from being uniformly distributed. To tackle the limitation of entropy maximization, we propose
(
α
,
β
)
-generalized KL divergence,
D
KL
α
,
β
(
p
‖
q
)
, which can be used to identify significantly more NC instances. Theoretical properties of
D
KL
α
,
β
(
p
‖
q
)
are proven, and we also show empirically that a simple use of
D
KL
α
,
β
(
p
‖
q
)
outperforms all baselines on the NC instance identification task. Building upon
(
α
,
β
)
-generalized KL divergence, we also introduce a new iterative training framework,
GenKL
, that identifies and relabels NC instances. When evaluated on three web image datasets, Clothing1M, Food101/Food101N, and mini WebVision 1.0, we achieved new state-of-the-art classification accuracies:
81.34
%
,
85.73
%
and
78.99
%
/
92.54
%
(top-1/top-5), respectively.