`

Timezone: »

 
Poster
Densely connected normalizing flows
Matej Grcić · Ivan Grubišić · Siniša Šegvić

Tue Dec 07 04:30 PM -- 06:00 PM (PST) @ Virtual #None

Normalizing flows are bijective mappings between inputs and latent representations with a fully factorized distribution. They are very attractive due to exact likelihood evaluation and efficient sampling. However, their effective capacity is often insufficient since the bijectivity constraint limits the model width. We address this issue by incrementally padding intermediate representations with noise. We precondition the noise in accordance with previous invertible units, which we describe as cross-unit coupling. Our invertible glow-like modules increase the model expressivity by fusing a densely connected block with Nyström self-attention. We refer to our architecture as DenseFlow since both cross-unit and intra-module couplings rely on dense connectivity. Experiments show significant improvements due to the proposed contributions and reveal state-of-the-art density estimation under moderate computing budgets.

Author Information

Matej Grcić (Faculty of Electrical Engineering and Computing, Unska 3, 10000 Zagreb, Croatia, VAT: HR57029260362)
Ivan Grubišić (UniZg-FER, University of Zagreb)
Siniša Šegvić (UniZg-FER)