Over the past several decades polar field precursor technique has been one of the important methods for predicting next solar cycle's strength. Since the success of predicting cycle 24 strength using this method, there now exists a general consensus that polar field from previous cycle is a good predictor for the amplitude of the next cycle. However, if one attempts to predict the next cycle's amplitude separately in North and South hemispheres from the polar field in each hemisphere, the prediction fails. This method fails on predicting asymmetries in both timing and amplitude between North and South hemispheres. Furthermore, we do not yet understand the physics behind the link between previous cycle's polar field and next sunspot cycle's amplitude. In this context, I will discuss the progress made so far to understand the physics behind different prediction methods and their limitations, and how the inclusion of data assimilation techniques can provide prospects for improvements. Equally important is predicting short-term or "seasonal" (6-18 months) solar activity bursts, their timing and amplitude, because most major space weather events originate from these bursts, and occur even in weak cycles. For example, a major Carrington type event occurred in cycle 24, which has been the weakest cycle in 100 years. The original Carrington event of 1859 also occurred in a weak cycle. But both occurred during the bursty season. Therefore along with progress in decadal solar activity prediction, the seasonal activity prediction is extremely important for avoiding hazardous effects from space weather. I will close by discussing the recent advances in seasonal activity prediction.
Mode of presentation: oral (Need to be confirmed by the SOC)