The deterministic information bottleneck

Lossy compression fundamentally involves a decision about what is relevant and what is not. The information bottleneck (IB) by Tishby, Pereira, and Bialek formalized this notion as an information-theoretic optimization problem and proposed an optimal tradeoff between throwing away as many bits as possible, and selectively keeping those that are most important. Here, we introduce an alternative formulation, the deterministic information bottleneck (DIB), that we argue better captures this notion of compression. As suggested by its name, the solution to the DIB problem is a deterministic encoder, as opposed to the stochastic encoder that is optimal under the IB.We then compare the IB and DIB on synthetic data, showing that the IB and DIB perform similarly in terms of the IB cost function, but that the DIB vastly outperforms the IB in terms of the DIB cost function. Moreover, the DIB offered a 1-2 order of magnitude speedup over the IB in our experiments. Our derivation of the DIB also offers a method for continuously interpolating between the soft clustering of the IB and the hard clustering of the DIB.