If by "active-active" you mean having running SQL Instances running secondary replicas of the database, then yes.
You'll need to carefully plan your cluster heartbeat settings, and your quorum model (see eg Cloud Quorum).
Because while it's technically possible to have automatic failover in this scenario, it's difficult. For most applications it's better to have a single async secondary in Azure and the primary and a synchronous secondary on-prem.
The difficulties are:
1) Only synchronous secondaries support automatic failover, and having a synchronous secondary in a remote datacenter can have a serious impact on performance.
And if you want automatic failover, you can't have a quorum voting majority in your datacenter; instead the quorum needs to be in the cloud, eg using Cloud Witness, which necessarily means:
2) In case of a temporary network issue between your Datacenter and Azure the nodes in your datacenter will go offline and the database will fail over.
And you'll need to carefully adjust the cluster timeout values which are always a tradeoff between failover time and false outages.