Since this comes up a lot, a quick run though of the testable, falsifiable, science that supports a human cause of recent trends in global mean temperature.

First off, we start with the observations:
1) spectra from space showing absorption of upward infra-red radiation from the Earth's surface.
2) Measurements from around the world showing increases in CO2, CH4, CFCs, N2O.
3) In situ & space based observations of land use change

We develop theories.
1) Radiative-transfer (e.g. Manabe and Wetherald, 1967)
2) Energy-balance models (Budyko 1961 and many subsequent papers)
3) GCMs (Phillips 1956, Hansen et al 1983, CMIP etc.)

We make falsifiable predictions. Here are just a few:
1967: increasing CO2 will cause the stratosphere to cool
1981: increasing CO2 will cause warming at surface to be detectable by 1990s
1988: warming from increasing GHGs will lead to increases in ocean heat content

1991: Eruption of Pinatubo will lead to ~2-3 yrs of cooling
2001: Increases in GHGs will be detectable in space-based spectra
2004: Increases in GHGs will lead to continued warming at ~0.18ºC/decade.

We test the predictions:
Stratospheric cooling?
Detectable warming?
OHC increase?
Pinatubo-related cooling?
Space-based changed in IR absorption?
post-2004 continued warming?

With this validated physics, we can estimate contributions to the longer term trends. https://www.bloomberg.com/graphics/2015-whats-warming-the-world/

This too is of course falsifiable. If one could find a model system that matches all of the previously successful predictions in hindcasts, and gives a different attribution, we could test that. [Note this does not (yet) exist, but let's keep an open mind].

We can also look at the testable, falsifiable, theories that were tested, and failed.
Solar forcing? Fails the strat cooling test.
Ocean circulation change? Fails the OHC increase test
Orbital forcing? Fails on multiple levels

If you have a theory that you don't think has been falsified, or you think you can falsify the mainstream conclusions, that's great! We can test that too! (But lots of people have tried this already so expect there to be an answer already).

PS. Actually, it's even a bit harder. Not only would you need to find a theory that does as well as the current theory, but you'd also need to show why the current theory isn't operative.