Abstract: This manuscript explores the potential of selflearning and multimodal agents within the financial sector. In particular, we discuss the multimodal agents' ability to process diverse data types and self-learning agents' ability to handle continuous improvement even without explicit programming. We also emphasize the various application areas where they are deployed along with the challenges. In addition, we explore practical considerations for their deployment and evaluation. We hypothesize that effective integration of these agents requires careful consideration of their impact in the real world and a balanced approach to autonomy, often incorporating human oversight.
External IDs:dblp:conf/bigdataconf/YadavLDGY25
Loading